Compute service#

The Compute Service provides the ability to compute the output of compute graphs.

OpenAPI file: ComputeService.json

Errors#

Errors shows how these errors are sent.

Each error’s errorName is Compute: followed by its name below.

Status	Name	Notes
400	`CannotAggregateEnumPlot`
400	`CannotCompileRegexp`	The regular expression could not be compiled.
400	`ChannelHasWrongType`
400	`ComputationTooComplex`	Computation too complex. Try reducing the number of channels, computations, or tags.
400	`ConverterInputUnitNotFound`	A unit could not be found for the input series, or it did not exist in the unit system. The conversion could not be computed.
400	`ConverterOutputUnitNotFound`	The output unit symbol could not be found in the unit system. The conversion could not be computed.
400	`CurveInvalidNegativeInputs`	Some points in the input had negative (or zero) x-values. Power and logarithmic curve fits do not allow negative (or zero) x-valued points.
400	`CurveUnequalInputLength`	The x and y inputs to fit a curve to do not have the same length.
400	`CustomCurveFormulaMissingParameter`	A custom curve fit declared a parameter (with an initial guess) that its formula never references. Every declared parameter must appear in the formula; otherwise it is an unconstrained degree of freedom for the solver. Remove the unused parameter or reference it in the formula.
400	`CustomCurveFormulaMissingVariable`	A custom curve fit declared a variable that its formula never references. The unused variable’s series would be fetched and aligned for nothing, so it almost certainly indicates a mistake in the formula. Remove the unused variable or reference it in the formula.
400	`CustomCurveMultipleTimeVariables`	A custom curve fit may declare at most one time variable; there is only one sample-timestamp axis for the formula to refer to.
400	`DerivedSeriesHasWrongType`
400	`DuplicateTimestamp`
400	`EmptyInput`	At least one input must be provided.
400	`EmptyPath`	Specified path cannot be empty.
400	`EmptySet`	The provided set must contain at least one element.
400	`ExponentialCurveInputTooLarge`	Due to limits in how numbers are represented by computers, exponential fits cannot be performed on data with x-values that are too large (> 650). This includes timestamp data based on the Time Unit parameter to the curve fit - e.g. 1 second at millisecond scale is too large, as it is 1000 milliseconds.
400	`FrequencyDomainNotEnoughData`	Not enough data points to estimate a sampling frequency for frequency-domain analysis. Try again with a larger time range.
400	`FrequencyDomainWindowEmpty`	The selected window for the frequency domain is empty. Try again with a larger time range.
400	`FrequencyDomainWindowTooLarge`	The selected window for the frequency domain is too large. Try again with a smaller time range.
400	`GranularityMismatch`	Some operations that require matching granularity did not have matching granularity.
400	`GroupByMissingTag`	Attempted to group by a tag that does not exist.
400	`GroupByTagsNotSubset`	Group by tags must be equal, or one set must be a subset of the other.
400	`GroupBysNotSupportedForQuery`	Group by queries are not supported for the compute node supplied
400	`InvalidBitOperation`	The bit operation is invalid due to the index or operand being out of range. Valid range - [0, 63].
400	`InvalidExpression`
400	`InvalidFieldName`	Struct field names cannot contain ‘.’
400	`InvalidLiteralRange`	The starting timestamp of a range cannot be greater than or equal to the end timestamp. At least one of start and end timestamps must be present. Only the first range when sorted may omit the start, and only the last may omit the end.
400	`InvalidNumericAggregations`	Numeric aggregations can only be specified on numeric series requests.
400	`InvalidNumericOutputFields`	Numeric output fields can only be specified on Arrow format numeric series requests. At least one field must be specified.
400	`InvalidPlotType`
400	`InvalidRangeNodeStartAfterViewRange`	The compute node was given a starting timestamp after the end of the view range window.
400	`InvalidRefpropInputCount`	REFPROP requires exactly two input series.
400	`InvalidSeriesLocator`
400	`InvalidTagFilterConfiguration`	Invalid tag filter configuration was provided.
400	`InvalidTimeUnit`	The time unit associated with the series is incompatible with the current transform.
400	`InvalidUdfSource`	The Rust UDF source failed validation or compilation.
400	`InvalidValueMap`	Range end must be greater than range start. Range start may only be undefined if in the first range, or the prior end is defined and less. Range end may only be undefined if in the last range, or the next start is defined and greater. Ranges must be in increasing order.
400	`MaxQuerySizeExceeded`	The maximum query size was exceeded.
400	`MemoryLimitExceeded`	The query exceeded the memory limit. Try querying a smaller time range or fewer series.
400	`MisalignedScatterPoints`	No scatter points could be aligned for plotting. This typically means the input series do not have overlapping timestamps within the configured interpolation window.
400	`MissingFunctionParameter`	The requested module function could not be executed because a required parameter was not provided.
400	`MissingModuleApplication`	The requested module application could not be found. This may be because the reference is incorrect or the application is unavailable in the current execution context.
400	`MissingModuleFunction`	The requested module function could not be found. This may be because the reference is incorrect or the function is unavailable in the current execution context.
400	`MissingModuleVariable`
400	`MissingVariable`
400	`NonPositiveDuration`	Duration must be strictly positive.
400	`NonPositiveResampleInterval`	The resample interval must be strictly positive.
400	`NotEnoughPointsForCurve`	There were not enough points to fit the curve requested.
400	`OverlappingLiteralRanges`	Ranges cannot overlap. Only the first range when sorted may omit the start, and only the last may omit the end.
400	`QueryRangeTooLarge`	The requested time range is too large. Try again with a smaller time range.
400	`RefpropInputTooLarge`	The number of points in the REFPROP inputs is too large.
400	`RefpropOutputPropertyIsInput`	The output property cannot be one of the input properties for REFPROP.
400	`RequestCancelled`	The compute request was cancelled by the user.
400	`RequestedRowsLimitExceeded`	The maximum number of points that can be returned is 10,000.
400	`ResampleDefaultValueTypeMismatchesSeriesType`	The default value provided for resampling does not match the type of the series being resampled.
400	`ResolutionIntervalTooSmallForRange`
400	`RowLimitExceeded`	The query exceeded that limit for number of rows returned. For the query to complete, either zoom in or downsample.
400	`S3ObjectStorageRateLimited`	S3 rejected the request with a TooManyRequests error. Try again with a smaller time range, fewer series, or after a short wait.
400	`SelectPropertyMissingProperty`	No resolved asset or run branch provides the selected property.
400	`SelectPropertyWrongType`	A resolved branch has the selected property, but its typed variant differs from the numeric or string type required by the series node.
400	`SelectSeriesNameConflictsWithChannelFilter`	A SelectSeries node provided both the `name` field and a NOMINAL_CHANNEL tag filter. Use the NOMINAL_CHANNEL filter alone to select channels.
400	`SelectSeriesResolvedNoChannels`	A channel selection (channel name plus tag filters) resolved to zero dataset channels for the requested scope and time range, so the series has nothing to read. Verify the channel exists in the dataset and that the tag filters match at least one channel over the requested range.
400	`SelectTagsNotSubset`	Select tags keys must be a subset of the tags available on the input series.
400	`ServerOverloaded`	The query was rejected because the compute backend is temporarily overloaded. This is a transient, load-shedding response that protects overall availability; the query itself is valid. Callers do not receive this error directly: the read path wraps it in a QoS unavailable response (503), so standard classic REST clients back off and retry automatically. The error type is retained as the 503’s cause for server-side logging and query_log correlation via queryId.
400	`SignalFilterInvalidCutoffBand`	Lower band cutoff frequency must be less than or equal to upper band cutoff frequency.
400	`SignalFilterInvalidCutoffFrequency`	Cutoff frequency must be less than half the sampling frequency.
400	`SignalFilterInvalidOrder`	Order must be positive even numbers for bidirectional filters. Bidirectional filters are applied once forwards and once backwards, effectively doubling order. Doubled again for bandpass/bandstop filters, requiring a multiple of 4 for filter order.
400	`SignalFilterNonPositiveCutoffFrequency`	Cutoff frequency must be strictly positive.
400	`SignalFilterNotEnoughData`	Not enough data to apply the filter. Required data scales with filter order. Try again with a larger time range.
400	`SourceBlocked`	Queries from this source are temporarily blocked.
400	`SpectrogramUnsupportedInput`	The spectrogram input reads from more than one data source or uses an operation spectrograms do not support.
400	`TimeoutExceeded`	The query timed out. Try querying a smaller time range or fewer series.
400	`TooFewInputs`	The operation requires more input series than were provided.
400	`TooManyBuckets`	The number of buckets returned is greater than the max allowed (1000). Try again by specifying a larger bucket width or fewer bucket counts.
400	`TooManyCategories`	An enum series has more categories than the max allowed.
400	`TooManyEvents`	Number of events is greater than the max allowed (1000).
400	`TooManyFrequencyBins`	The number of one-sided frequency bins exceeds the maximum allowed. Reduce nfft in transform settings to produce fewer bins.
400	`TooManyLiterals`	Too many literals provided in a literal node.
400	`TooManyPoints`	The number of points returned is greater than the max allowed. Try again with a smaller time range, or a larger resolution.
400	`TooManyRanges`	Ranges is greater than the max allowed (2000).
400	`TooManyResamplePoints`	The resample interval is too short and would produce more than the max points (10,000,000). Try again by specifying a larger interval.
400	`TooManySearchResults`	Asset or run search returned more matches than a single page allows. Refine the query or raise maxResults for the search frame so all matches fit in one page.
400	`TooManySubrequests`	Number of parameterized subrequests is greater than the max allowed (300).
400	`TooManyValueMapRanges`	A value map series contains too many range mappings.
400	`TruncationNotSupportedForSeriesType`	The requested series type does not support truncation for this compute request.
400	`UdfExecutionLimitExceeded`	A user-defined function exceeded its per-query execution limit while processing this query. Try querying a smaller time range, increasing the aggregation bucket size to reduce the number of UDF invocations, or simplifying the function’s per-row logic.
400	`UdfMemoryLimitExceeded`	A user-defined function exceeded its per-invocation memory limit.
400	`UdfRuntimeError`	A user-defined function panicked or aborted at runtime.
400	`UnequalInputLength`	The paired inputs for a calculation do not have the same length.
400	`UnitConversionInvalid`	The unit conversion could not be computed because the input unit cannot be converted to the output unit.
400	`UnknownBuiltinConstant`
400	`UnsupportedRefpropProperty`	The specified REFPROP property is not supported or not available in the current configuration.
400	`UnsupportedRefpropSubstance`	The specified REFPROP substance is not supported or not available in the current configuration.
400	`VariableHasWrongType`
403	`NotAuthorized`
500	`ComputeBackendUnavailable`	The compute backend was temporarily unavailable before the query could be submitted. Retry the request after a short delay.
500	`ConcurrentQueriesExceeded`	There are too many concurrent queries running.
500	`UdfNotRegistered`	A native user-defined function was not registered on the when the query executed. This indicates an internal failure rather than invalid user input.
500	`UnexpectedTimeout`	An unexpected timeout occurred when executing a query. This indicates an internal error where the query may have succeeded but the response did not reach the client.

Each error’s errorName is ExternalDatabase: followed by its name below.

Status	Name	Notes
400	`ExternalDatabaseGatewayTimeout`	A gateway timeout was encountered while trying to fetch data from the external database.
400	`ExternalDatabaseRateLimitExceeded`	The external database rate limit was exceeded.
500	`ExternalDatabaseBadGateway`	A bad gateway result was encountered while trying to fetch data from the external database.
500	`ExternalDatabaseSocketTimeout`	A socket timeout was encountered while trying to fetch data from the external database.

Types#

Abs#

Computes the absolute value of each point in the series.

x NumericSeries required: Input numeric series.

AbsoluteThreshold#

Threshold defined as a real number corresponding the unit of a series.

value DoubleConstant required: Allowed spread, in the series’ own unit.

AbsoluteTimestampSeries#

Replaces values in a series with the absolute timestamps.

input Series required
timeUnit TimeUnit required: The time unit used to define the output values. Throws an error if nanosecond or finer.

Acos#

Computes the arccosine of each value, returning radians.

x NumericSeries required: Input numeric series (values in [-1, 1]).

Add#

Adds two numeric series pointwise.

left NumericSeries required: Left-hand operand.
right NumericSeries required: Right-hand operand.

AfterPersistenceWindow#

The first point in the output range will be the first point after the condition has been true for the min duration and min points.

An empty JSON object, {}.

AggregateEnumSeries#

Aggregates values with duplicate timestamps in the input series values into a single value using the specified aggregation function.

input EnumSeries required
function EnumAggregationFunction required

AggregateNumericSeries#

DEPRECATED. Use a dedicated group-by-and-aggregate nodes instead.

Aggregates values with duplicate timestamps in the input series values into a single value using the specified aggregation function.

input NumericSeries required
function NumericAggregationFunction required
groupByTags array of StringConstant: Tags to group by for the aggregation. If left empty, the tags to group by will be equivalent to those in the input series. If specified, the result will be grouped ONLY by the specified tags. The tags specified here MUST be a (non-strict) subset of the input series’s group by tags.
aggregateByAllGroupings boolean: This field’s purpose is to distinguish between the two flavors of groupByTags being empty. It has no effect when groupByTags is non-empty. When true and groupByTags is empty, aggregate across all series in the input. When false and groupByTags is empty, the result will be grouped by the same tags as the input series.
interpolationConfiguration InterpolationConfiguration deprecated: If provided, interpolates values at timestamps where the input series has values before aggregating. If not provided, only aggregates when timestamps match exactly (existing behavior). Deprecated: Setting interpolation on this node is deprecated.

AggregateUnderRangesSeries#

Aggregates the input series under the ranges in the provided ranges series, and outputs a new series with one point at the start of each range, where the value is the aggregation result. If a range has no points, no point will be produced for it. If a range has no start, the point produced for it will be at the start of the compute request range.

input NumericSeries required
ranges RangeSeries required
operation NumericAggregationFunction required deprecated: Deprecated: Use a dedicated group-by-and-aggregate node instead (coming in a follow-up PR). This field is retained as required for backwards compatibility with existing callers.

AggregationBuilder#

Set type to the name of one of the other fields, and set only that field.

rollingTime RollingTimeAggregationBuilder
rollingPoints RollingPointsAggregationBuilder
groupBy GroupByAggregationBuilder

Alignment#

Specifies how to set the timestamp grid for multi-input operations.

Set type to the name of one of the other fields, and set only that field.

driverSeries DriverSeries
union Union

AlignmentConfiguration#

interpolationConfiguration InterpolationConfiguration required: The interpolation rules to use when aligning the non-driver series.
driverSeries AlignmentDriverSeries required: The series whose timestamps drive alignment.

AlignmentDriverSeries#

FIRST
SECOND

AllowNegativeValues#

An empty JSON object, {}.

AnchorAlignedDataset#

Aligns each anchored portion of the input dataset onto a common anchor timestamp while preserving grouping tags. Without a reference the selected anchor becomes timestamp zero, so the output is in a relative time domain; with a reference the anchors are aligned onto the referenced portion’s anchor and the output stays in the input’s absolute time domain. Currently run boundary, event, and range anchors are supported, and every storage-backed leaf of the input must resolve to run-backed branches.

input Dataset required: The dataset to align.
anchor DatasetAnchor required: Selects the timestamp each anchored portion of the input is aligned by.
reference AnchorReference: When present, each anchor is aligned onto the referenced anchor timestamp instead of onto timestamp zero, so output remains in the input’s absolute time domain and the referenced portion is unshifted. When absent, the selected anchor becomes timestamp zero.

AnchorReference#

Selects the anchor timestamp every other anchor is aligned onto.

Set type to the name of one of the other fields, and set only that field.

earliestAnchor EarliestAnchorReference: Aligns every anchor onto the earliest anchor resolved from the input, so the earliest-anchored run is unshifted and every other run shifts backwards onto it.

AndSeries#

Combines two boolean series point-wise using logical AND, producing true where both inputs are true and false otherwise.

left BooleanSeries required
right BooleanSeries required

ApproximateThresholdOperator#

EQUAL_TO
NOT_EQUAL_TO

ApproximateThresholdRanges#

Produces a list of ranges for which the threshold condition is satisfied.

input NumericSeries required
threshold DoubleConstant required
tolerance DoubleConstant required
operator ApproximateThresholdOperator required
persistenceWindowConfiguration PersistenceWindowConfiguration

ArithmeticSeries#

inputs map of LocalVariableName to NumericSeries
expression string required
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

ArraySeries#

Set type to the name of one of the other fields, and set only that field.

numeric1d Numeric1dArraySeries
enum1d Enum1dArraySeries

ArrowArrayPlot#

Set type to the name of one of the other fields, and set only that field.

bucketedNumeric BucketedNumericArrayPlot
bucketedEnum BucketedEnumArrayPlot

ArrowPlot#

A generic Arrow IPC plot response. Used for all Arrow-serialized plot types (numeric, bucketed numeric, enum, bucketed enum, struct, bucketed struct, full resolution, and bucketed multivariate).

arrowBinary string (base64) required: The raw binary containing an Arrow IPC stream.
groupByKeys array of string: This field specifies the tags that the final output is grouped by. When you combine multiple channels, this list represents the superset of all group by keys used across every individual channel.
nextPageToken PageToken: Continuation token for this paged result. Can only be returned if paging was requested. To retrieve the next page, repeat the same request with this value as pageToken and keep the pageSize sign unchanged. Returned only when this response contains exactly abs(pageSize) points; the next response may still be empty if this page ended exactly at the result-set boundary. Empty means this response was short and there are no further points in the requested direction.

Asin#

Computes the arcsine of each value, returning radians.

x NumericSeries required: Input numeric series (values in [-1, 1]).

Asset#

Asset reference for Dataset.asset. The rid field accepts either a literal asset RID or a StringConstant reference resolved from Context.

rid StringConstant required: Resource identifier of an asset

AssetChannel#

assetRid StringConstant required
dataScopeName StringConstant required: Used to disambiguate when multiple data scopes within this asset contain channels with the same name.
channel StringConstant required
additionalTags map of string to StringConstant deprecated: Deprecated: Use additionalTagFilters
additionalTagFilters TagFilters: Tags to filter the channel by, in addition to tag filters defined for a given Asset data scope. Throws on collisions with tag keys already defined for the given Asset data scope. Only returns points that match both sets of tag filters. For log series, include arg filters here in addition to tag filters.
tagsToGroupBy array of string deprecated: Deprecated: Use groupByTags instead. Throws if both tagsToGroupBy and groupByTags are specified.
groupByTags array of StringConstant: Tags that the channel should be grouped by. If this is non-empty a grouped result will be returned with an entry for each grouping. Only one of tagsToGroupBy and groupByTags should be specified.

AssetsSearchTarget#

Search assets. Each match expands to one grouping per data scope on that asset.

query ResourceSearchQuery required: Predicate tree used to filter matching assets.

Atan2#

Computes the two-argument arctangent atan2(y, x) using signs to determine the quadrant.

y NumericSeries required: Y-coordinate series.
x NumericSeries required: X-coordinate series.

Average#

A average that gives the same weight to each observation inside the time window

An empty JSON object, {}.

BandPassConfiguration#

lowPassbandFrequency DoubleConstant required
highPassbandFrequency DoubleConstant required

BandStopConfiguration#

lowStopbandFrequency DoubleConstant required
highStopbandFrequency DoubleConstant required

BinaryArithmeticOperation#

ATAN2: Calculates the angle measure between the x-axis and a ray from the origin to a point (input2, input1) in radians. Note the order of arguments.

BinaryArithmeticSeries#

Applies a point-wise transformation to a pair of series.

input1 NumericSeries required
input2 NumericSeries required
operation BinaryArithmeticOperation required
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

BitAnd#

Applies a bitwise AND between each value (cast to int64) and the operand.

input NumericSeries required: Input numeric series (values cast to int64).
operand IntegerConstant required: The value to AND with each point.

BitAndFunction#

operand integer (int64) required

BitOperationFunction#

Set type to the name of one of the other fields, and set only that field.

and BitAndFunction
or BitOrFunction
xor BitXorFunction
shiftRight BitShiftRightFunction
shiftLeft BitShiftLeftFunction
bitTest BitTestFunction

BitOperationSeries#

Casts input series values to long before applying the bitwise operation.

input NumericSeries required
function BitOperationFunction required

BitOr#

Applies a bitwise OR between each value (cast to int64) and the operand.

input NumericSeries required: Input numeric series (values cast to int64).
operand IntegerConstant required: The value to OR with each point.

BitOrFunction#

operand integer (int64) required

BitShiftLeft#

Left-shifts the bits in each value (cast to int64) by the given number of positions.

input NumericSeries required: Input numeric series (values cast to int64).
operand IntegerConstant required: Number of positions to shift left.

BitShiftLeftFunction#

Shifts the bits in each value left according to the given operand.

operand integer required

BitShiftRight#

Right-shifts the bits in each value (cast to int64) by the given number of positions.

input NumericSeries required: Input numeric series (values cast to int64).
operand IntegerConstant required: Number of positions to shift right.

BitShiftRightFunction#

Right shifts the bits in each value right according to the given operand.

operand integer required

BitTest#

Returns the bit at the specified index for each value (cast to int64), where index 0 is the least significant bit.

input NumericSeries required: Input numeric series (values cast to int64).
index IntegerConstant required: Zero-based bit index, where 0 is the least significant bit.

BitTestFunction#

Returns the bit at the specified index, where the right-most bit has index 0.

index integer required

BitXor#

Applies a bitwise XOR between each value (cast to int64) and the operand.

input NumericSeries required: Input numeric series (values cast to int64).
operand IntegerConstant required: The value to XOR with each point.

BitXorFunction#

operand integer (int64) required

Bode#

Returns the bode magnitude and phase of a system’s frequency response.

input NumericSeries required
output NumericSeries required
alignmentConfiguration AlignmentConfiguration: When present, aligns one Bode input to the other using the configured interpolation and driver series before estimating the frequency response. When absent, Bode preserves the existing strict behavior and requires the input and output value arrays to have equal length. For Bode, FIRST uses the input timestamps and SECOND uses the output timestamps.
stftOptions StftOptions
magnitudeScaling MagnitudeScaling: The scaling to apply to the output magnitude. Defaults to MAGNITUDE_DB_20 if not specified.
outputFrequencyType OutputFrequencyType: The type of the output frequency. Defaults to LINEAR if not specified.
unwrapPhase boolean: Unwrap the phase of the output. Defaults to true if not specified.

BooleanRanges#

Produces ranges spanning the regions where a boolean input is true.

The condition is evaluated point-wise at each aligned timestamp; the numeric series it compares are aligned against each other the same way arithmetic between them would be.

input BooleanSeries required: The condition whose true regions become ranges.
minPoints IntegerConstant: Minimum number of matching points. Defaults to 1.
minDuration compute.api.Duration: Minimum matching duration. Defaults to one nanosecond.
rangeStart RangeStart: Selects whether the output begins at the first match or after persistence is satisfied.

BooleanSeries#

Set type to the name of one of the other fields, and set only that field.

greaterThan GreaterThanSeries
lessThan LessThanSeries
equalTo EqualToSeries
notEqualTo NotEqualToSeries
greaterThanOrEqualTo GreaterThanOrEqualToSeries
lessThanOrEqualTo LessThanOrEqualToSeries
and AndSeries
or OrSeries

BucketedCartesian3dPlot#

buckets array of Cartesian3dBucket

BucketedCartesianPlot#

buckets array of CartesianBucket

BucketedEnumArrayPlot#

The array is flattened out into a an arrow stream of bucketed primitive results, with an extra column to indicate the index of the array that the bucket corresponds to.

arrowBinary string (base64) required: The raw binary containing Arrow IPC stream for a bucketed N-dimensional enum array plot.
groupByKeys array of string: This field specifies the tags that the final output is grouped by. When you combine multiple channels, this list represents the superset of all group by keys used across every individual channel.

BucketedEnumPlot#

timestamps array of Timestamp: The end of the bucket, exclusive.
buckets array of EnumBucket
categories array of string

BucketedFrequencyDomainPlot#

A bucketed/decimated frequency domain plot where multiple frequency bins have been aggregated.

buckets array of FrequencyBucket

BucketedMultivariatePlot#

buckets array of MultivariateBucket
groupByKeys array of string: This field specifies the tags that the final output is grouped by. When you combine multiple channels, this list represents the superset of all group by keys used across every individual channel.

BucketedNumericArrayPlot#

The array is flattened out into a an arrow stream of bucketed primitive results, with an extra column to indicate the index of the array that the bucket corresponds to.

arrowBinary string (base64) required: The raw binary containing Arrow IPC stream for a bucketed N-dimensional numeric array plot.
groupByKeys array of string: This field specifies the tags that the final output is grouped by. When you combine multiple channels, this list represents the superset of all group by keys used across every individual channel.

BucketedNumericPlot#

timestamps array of Timestamp: The end of the bucket, exclusive.
buckets array of NumericBucket

BuiltinConstant#

A system-defined constant referenceable by name. The server resolves the symbol to its numeric value during query resolution.

Set type to the name of one of the other fields, and set only that field.

math MathConstant
physical PhysicalConstant

BuiltInCurveFit#

A closed-form curve fit (exponential, logarithmic, or power) of the enclosing target (y) against the chosen independent variable (x): either sample time (a time variable) or another series (a numeric variable).

independentVariable CurveFitVariable required: The x axis the target is fit against — either relative sample time or another series.

Cartesian#

Set type to the name of one of the other fields, and set only that field.

scatter Scatter

Cartesian3d#

Set type to the name of one of the other fields, and set only that field.

scatter3d Scatter3d

Cartesian3dBounds#

Min/max bounds of an XYZ Cartesian plot, inclusive.

minX DoubleConstant required
maxX DoubleConstant required
minY DoubleConstant required
maxY DoubleConstant required
minZ DoubleConstant required
maxZ DoubleConstant required

Cartesian3dBucket#

minX number deprecated: Deprecated: Switching to point-based min/maxes
maxX number deprecated: Deprecated: Switching to point-based min/maxes
meanX number required
minY number deprecated: Deprecated: Switching to point-based min/maxes
maxY number deprecated: Deprecated: Switching to point-based min/maxes
meanY number required
minZ number deprecated: Deprecated: Switching to point-based min/maxes
maxZ number deprecated: Deprecated: Switching to point-based min/maxes
meanZ number required
minXPoint Point3d required
maxXPoint Point3d required
minYPoint Point3d required
maxYPoint Point3d required
minZPoint Point3d required
maxZPoint Point3d required
minTimestamp Timestamp required
maxTimestamp Timestamp required
count integer (int64) required

Cartesian3dUnitResult#

x UnitResult required
y UnitResult required
z UnitResult required

CartesianBounds#

Min/max bounds of an XY Cartesian plot, inclusive.

minX DoubleConstant required
maxX DoubleConstant required
minY DoubleConstant required
maxY DoubleConstant required

CartesianBucket#

minX number required
maxX number required
minY number required
maxY number required
minTimestamp Timestamp required
maxTimestamp Timestamp required
count integer (int64) required

CartesianPlot#

timestamps array of Timestamp
xValues array of number
yValues array of number

CartesianUnitResult#

x UnitResult required
y UnitResult required

CategoricalEqualityRangesV2#

Produces ranges where all categorical inputs are equal or are not all equal.

inputs array of EnumSeries: The series compared against each other. At least two are required.
minPoints IntegerConstant: Minimum number of matching points. Defaults to 1.
minDuration compute.api.Duration: Minimum matching duration. Defaults to one nanosecond.
rangeStart RangeStart: Selects whether the output begins at the first match or after persistence is satisfied.
equalityOperator EqualityOperator required

CategoricalOnChangeRangesV2#

Produces a zero-duration range whenever a categorical input changes value.

input EnumSeries required

CategoricalStaleRangesV2#

Produces ranges where a categorical input has no data for at least a duration.

input EnumSeries required
threshold compute.api.Duration required: Minimum gap without data that produces an output range.

CategoryFilterRangesV2#

Produces ranges where a categorical input matches a set of values.

input EnumSeries required
values array of StringConstant: Category labels matched against each input point.
minPoints IntegerConstant: Minimum number of matching points. Defaults to 1.
minDuration compute.api.Duration: Minimum matching duration. Defaults to one nanosecond.
rangeStart RangeStart: Selects whether the output begins at the first match or after persistence is satisfied.
operator EnumFilterOperator required

ChannelSeries#

Deprecated: prefer SelectSeries-based nodes with an explicit dataset (or VariableValue.series when binding a series to a variable), which select from an arbitrary dataset and let the consuming node decide the series type.

Set type to the name of one of the other fields, and set only that field.

dataSource DataSourceChannel
asset AssetChannel
run RunChannel

CombinedDataset#

Combines multiple datasets into a single dataset by taking the union of all points from each input (akin to SQL {@code UNION ALL}). Duplicate timestamps are preserved rather than merged; callers that need aggregation should wrap the selected series downstream.

inputs array of Dataset: The datasets to combine.

CombineStructSeries#

Combines named input series into one struct-valued series. Each map key becomes a field name in the resulting struct; a downstream UDF receives the struct as a JSON object keyed by these names and parses it itself. Alignment uses the interpolation configuration before producing row-aligned values.

inputs map of string to Series: Input series keyed by struct field name.
interpolationConfiguration InterpolationConfiguration deprecated: Optional interpolation rules used when aligning input series before combining them. Deprecated: Setting interpolation on this node is deprecated.

CompactEnumPoint#

timestamp Timestamp required
value integer required

ComputableNode#

Set type to the name of one of the other fields, and set only that field.

ranges SummarizeRanges
series SummarizeSeries
value SelectValue
cartesian SummarizeCartesian
cartesian3d SummarizeCartesian3d
frequency FrequencyDomain
frequencyV2 FrequencyDomainV2
spectrogram Spectrogram
histogram Histogram
curveV2 CurveFitV2
curve CurveFit
multivariate SummarizeMultivariate

compute.api.Duration#

A length of time. Can be constructed from literal values in various units or composed via arithmetic. Used for window sizes, time shifts, persistence thresholds, and resampling intervals.

Set type to the name of one of the other fields, and set only that field.

reference Reference
nanoseconds DurationNanoseconds
milliseconds DurationMilliseconds
seconds DurationSeconds
minutes DurationMinutes
hours DurationHours
days DurationDays
add DurationAdd
subtract DurationSubtract
negate DurationNegate

ComputeEventQuery#

Set type to the name of one of the other fields, and set only that field.

searchText StringConstant
after TimestampConstant: Filters to events after this timestamp, exclusive. This includes events that start before, but end after this time.
before TimestampConstant: Filters to events before this timestamp, exclusive. This includes events that start before, but end after this time.
advancedTimeFilter EventTimeFilter
asset StringConstant
template StringConstant
workbook StringConstant
dataReview StringConstant
originType SearchEventOriginType
dataReviewCheck StringConstant
dispositionStatus EventDispositionStatus
priority StringConstant
assignee StringConstant
eventType EventType
createdBy StringConstant
label StringConstant
property Property
and array of ComputeEventQuery
or array of ComputeEventQuery
not ComputeEventQuery
workspace StringConstant
procedure StringConstant
procedureExecution StringConstant
stepId StringConstant

ComputeNode#

Set type to the name of one of the other fields, and set only that field.

enum EnumSeries
numeric NumericSeries
log LogSeries
ranges RangeSeries
array ArraySeries
struct StructSeries
curveFit CurveFit: Deprecated: use curveV2 (CurveFitV2) instead.
curveV2 CurveFitV2
video Reference
raw Reference deprecated: Deprecated: wrap raw references in a typed compute node.

ComputeNodeRequest#

node ComputableNode required
start Timestamp required: The start of the time range (inclusive).
end Timestamp required: The end of the time range (inclusive).
context Context required
sourceRid string (RID): Optional RID identifying the resource that initiated this query (e.g. workbook/notebook RID, checklist RID, dataset RID). Used for observability only — trusted as-is, no permission checks are performed on this value.
requestId string (UUID): Optional UUID to provide as a handle to selectively cancel the execution of this compute request.

ComputeNodeResponse#

Set type to the name of one of the other fields, and set only that field.

range array of Range
rangesSummary RangesSummary
rangeValue Range
numeric NumericPlot
bucketedNumeric BucketedNumericPlot
numericPoint NumericPoint
singlePoint SinglePoint
arrowNumeric ArrowPlot
arrowBucketedNumeric ArrowPlot
enum EnumPlot
enumPoint EnumPoint
bucketedEnum BucketedEnumPlot
arrowEnum ArrowPlot
arrowBucketedEnum ArrowPlot
pagedLog PagedLogPlot
logPoint LogPoint
arrowLog ArrowPlot
cartesian CartesianPlot
bucketedCartesian BucketedCartesianPlot
bucketedCartesian3d BucketedCartesian3dPlot
frequencyDomain FrequencyDomainPlot
frequencyDomainV2 FrequencyDomainPlotV2
spectrogram SpectrogramPlot
bucketedFrequencyDomain BucketedFrequencyDomainPlot
numericHistogram NumericHistogramPlot
enumHistogram EnumHistogramPlot
curveFit CurveFitResult: Deprecated: use curveFitV2 (CurveFitResultV2) instead. Returned only for deprecated CurveFit requests.
curveFitV2 CurveFitResultV2
grouped GroupedComputeNodeResponses
array ArrowArrayPlot
bucketedStruct ArrowPlot
arrowStruct ArrowPlot
fullResolution ArrowPlot
arrowBucketedMultivariate ArrowPlot
multivariate BucketedMultivariatePlot

ComputeNodeResult#

Set type to the name of one of the other fields, and set only that field.

success ComputeNodeResponse
error ErrorResult

ComputeNodeWithContext#

seriesNode ComputeNode required
context Context required

ComputeUnitResult#

Set type to the name of one of the other fields, and set only that field.

single UnitResult
cartesian CartesianUnitResult
cartesian3d Cartesian3dUnitResult
multivariate MultivariateUnitResult

ComputeUnitsRequest#

node ComputableNode required
context Context required

ComputeWithUnitsResult#

computeResult ComputeNodeResult required
unitResult ComputeUnitResult required

ConstantNumericSeries#

A synthetic series that produces a constant value across the entire time range.

Set type to the name of one of the other fields, and set only that field.

double DoubleConstant: Create from a floating-point constant.

ContainsSeries#

Maps each log line to 1.0 when operator matches it and 0.0 otherwise.

input LogSeries required
operator LogFilterOperator required

Context#

variables map of VariableName to VariableValue
functionVariables map of FunctionReference to FunctionVariables deprecated: Deprecated: This field is deprecated and will be removed in a future version.
datasetReferences map of DatasetReferenceName to Dataset: Named datasets looked up by DatasetReference nodes in the compute tree.
defaultFillStrategy FillStrategy: Default fill strategy applied at every multi-input alignment boundary where no per-node interpolationConfiguration is specified. When absent, falls back to the server-configured default (forward fill with a 1 second fill limit).
defaultAlignmentStrategy Alignment: Default alignment strategy applied at every multi-input alignment boundary where no per-node alignment strategy is specified. When absent, falls back to the server default (driver-series alignment).
applyDataScopeOffsets boolean: When true, dataset resolution applies each data scope’s persisted offset as a time shift, and the client must not pre-apply those offsets itself. When absent or false, scope offsets are left to the client, preserving behavior for legacy clients that inject equivalent timeShift nodes into the compute request. Read from the context in scope at resolution time.
resolveAssetsThroughDatasets boolean: When true, Dataset.asset, the asset-typed sources of Dataset.run, and asset or run search expand through each asset’s mirrored derived dataset instead of its data scopes. Branches then carry the dataset’s declared channel-search split tag keys in place of the reserved DATA_SCOPE tag, so callers must filter and group on those keys; the definition’s own time shifts apply regardless of applyDataScopeOffsets; and an asset with no mirrored dataset resolves to no data. When absent or false, assets expand from their data scopes as before.

Cos#

Computes the trigonometric cosine of each value (input in radians).

x NumericSeries required: Input numeric series (radians).

Count#

The number of points inside the time window.

An empty JSON object, {}.

Cpsd#

Returns the cross-spectral density magnitude and phase of the two input series.

x NumericSeries required
y NumericSeries required
stftOptions StftOptions
magnitudeScaling MagnitudeScaling: The scaling to apply to the output magnitude. Defaults to MAGNITUDE_DB_10 if not specified.
outputFrequencyType OutputFrequencyType: The type of the output frequency. Defaults to LINEAR if not specified. Changing the output frequency type may also rescale the magnitude of the output in order to ensure the density of the output is consistent.
unwrapPhase boolean: Unwrap the phase of the output. Defaults to true if not specified.
outputPhaseUnit OutputPhaseUnit: The unit of the output phase. Defaults to RADIANS if not specified.

CumulativeSumSeries#

Calculates the running total of the series values.

input NumericSeries required
startTimestamp TimestampConstant required

CurveFit#

Deprecated: use CurveFitV2 instead.

curveFitPlotType CurveFitPlotType required
curveFitDetails CurveFitDetails required
curveFitOptions CurveFitOptions required

CurveFitBoundsConfiguration#

min DoubleConstant
max DoubleConstant

CurveFitDetails#

Set type to the name of one of the other fields, and set only that field.

exponential ExponentialCurve
logarithmic LogarithmicCurve
polynomial PolynomialCurve
power PowerCurve

CurveFitOptions#

startTime TimestampConstant required
endTime TimestampConstant required

CurveFitPlotType#

Set type to the name of one of the other fields, and set only that field.

timeSeries TimeSeriesCurveFit
scatter ScatterCurveFit

CurveFitRange#

startTime TimestampConstant required
endTime TimestampConstant required

CurveFitResult#

r2 number required: R^2 (coefficient of determination) for the fit curve, a normalized measure of how well the curve fits the data. Usually ranges from 0 to 1, with higher indicating better fit (points closer to fit line).
mae number required: Mean Absolute Error (MAE) - the average of the absolute differences between predicted and actual values. Measured in the same units as the y-values.
mse number required: Mean Squared Error (MSE) - the average of the squared differences between predicted and actual values. Measured in the squared units of the y-values.
rmse number required: Root Mean Squared Error (RMSE) - the square root of the mean squared error. Measured in the same units as the y-values.
maxAbsoluteError number required: Maximum Absolute Error - the largest absolute difference between any predicted and actual value. Measured in the same units as the y-values.
sampleCount integer required: The number of data points used in the curve fitting calculation.
curveResultDetails CurveResultDetails required: Description of the fit curve.

CurveFitResultV2#

r2 number required: R^2 (coefficient of determination) for the fit curve, a normalized measure of how well the curve fits the data. Usually ranges from 0 to 1, with higher indicating better fit (points closer to fit line).
mae number required: Mean Absolute Error (MAE) - the average of the absolute differences between predicted and actual values. Measured in the same units as the y-values.
mse number required: Mean Squared Error (MSE) - the average of the squared differences between predicted and actual values. Measured in the squared units of the y-values.
rmse number required: Root Mean Squared Error (RMSE) - the square root of the mean squared error. Measured in the same units as the y-values.
maxAbsoluteError number required: Maximum Absolute Error - the largest absolute difference between any predicted and actual value. Measured in the same units as the y-values.
sampleCount integer required: The number of data points used in the curve fitting calculation.
curveResultDetails CurveResultDetailsV2 required: Description of the fit curve.

CurveFitSeriesVariable#

series NumericSeries required
boundsConfig CurveFitBoundsConfiguration: Optional inclusive bounds selecting which variable samples are fit. Samples outside the bounds are excluded.

CurveFitTarget#

The series a curve is fit against (the y of the fit). Modeled as a union so additional target types can be added in the future; today the only arm is a numeric series.

Set type to the name of one of the other fields, and set only that field.

numeric CurveFitSeriesVariable

CurveFitTimeVariable#

timeUnit TimeUnit: The unit of the time variable in the formula expression. If not specified, the time unit is assumed to be seconds.

CurveFitV2#

target CurveFitTarget required: Target to fit the equation against.
fitRange CurveFitRange required: The range to fit the curve over
curveToFit CurveFitV2Details required: The curve type to fit against the target series.

CurveFitV2Details#

Set type to the name of one of the other fields, and set only that field.

exponential BuiltInCurveFit
logarithmic BuiltInCurveFit
polynomial PolynomialCurveFit
power BuiltInCurveFit
custom CustomCurveFit

CurveFitVariable#

A series a curve fit reads from: another numeric series, or the target’s own sample timestamps. Used both as the named formula inputs of a custom fit and as the independent (x) axis of a built-in fit.

Set type to the name of one of the other fields, and set only that field.

numeric CurveFitSeriesVariable
time CurveFitTimeVariable

CurveResultDetails#

Set type to the name of one of the other fields, and set only that field.

exponential ExponentialResultDetails
logarithmic LogarithmicResultDetails
polynomial PolynomialResultDetails
power PowerResultDetails

CurveResultDetailsV2#

Set type to the name of one of the other fields, and set only that field.

exponential ExponentialResultDetails
logarithmic LogarithmicResultDetails
polynomial PolynomialResultDetails
power PowerResultDetails
custom CustomCurveResultDetails

CustomCurveFit#

formula StringConstant required: A mathematical expression in terms of variable and parameter names, e.g. a * x + b * sin(x). Variable names must match keys in the variables map; parameter names must match keys in the parameters map.
variables map of LocalVariableName to CurveFitVariable: Named input series referenced in the formula. Keys must match the variable names used in the formula expression.
parameters map of string to DoubleConstant: Named fitting parameters and their initial guesses. Keys must match the parameter names used in the formula expression; values are the starting points passed to the optimizer. The target the curve is fit against is the y of the enclosing curve fit (the timeSeries series values, or the scatter y series).

CustomCurveResultDetails#

The result of fitting a custom curve, containing the optimized values for each named parameter.

parameters map of string to number: Optimized parameter values produced by the curve fit. Keys match the parameter names provided in the request.
converged boolean required: Whether the optimizer converged to a solution within its iteration limit. When false, the returned parameters are the best estimate found before the iteration cap was reached and did not satisfy the solver’s convergence tolerances.
iterations integer required: Number of iterations the solver completed.

Dataset#

A logical bundle of named series. For the time being, columns are each a time series (series name + tags). A dataset is either rooted in persisted data (asset or dataset) or defined as a transformation of another dataset.

Set type to the name of one of the other fields, and set only that field.

asset Asset: Single asset. Expands to one grouping per data scope on the asset. Groupings are tagged by assetRid and dataScope so callers can safely compose multiple assets with combine.
run Run: Single run. Expands to one grouping per data source attached to the run: asset-typed sources contribute one grouping per data scope on the asset; dataset-typed sources contribute one grouping per dataset. Groupings are tagged by runRid and dataScope; asset-typed groupings additionally carry assetRid so callers can safely compose multiple runs with combine.
saved SavedDataset
search SearchDataset
combine CombinedDataset
tag TaggedDataset
filter FilteredDataset
timeShift TimeShiftedDataset
alignByAnchor AnchorAlignedDataset
withSeries WithSeriesDataset
dropSeries DropSeriesDataset
reference DatasetReference
withDimensions DimensionsDataset

DatasetAnchor#

Set type to the name of one of the other fields, and set only that field.

run IntervalAnchor: Uses the start or end of each run as its anchor. The input dataset must resolve exclusively to run-backed branches.

DatasetReference#

References a named dataset in Context.datasetReferences.

name StringConstant required: Name of the dataset reference to resolve from the context.

DataSourceAndChannel#

dataSourceRid string (RID) required
channel string required

DataSourceChannel#

dataSourceRid StringConstant required
channel StringConstant required
tags map of string to StringConstant deprecated: Deprecated: Use tagFilters
tagFilters TagFilters: Tags to filter the channel by. Only returns points from the channel where tag values match the provided expression. For log series, include arg filters here in addition to tag filters.
tagsToGroupBy array of string deprecated: Deprecated: Use groupByTags instead. Throws if both tagsToGroupBy and groupByTags are specified.
groupByTags array of StringConstant: Tags that the channel should be grouped by. If this is non-empty a grouped result will be returned with an entry for each grouping. Only one of tagsToGroupBy and groupByTags should be specified.

DecimateStrategy#

Set type to the name of one of the other fields, and set only that field.

resolution DecimateWithResolution
buckets DecimateWithBuckets

DecimateWithBuckets#

buckets integer required: Number of points to generate in the output series.

DecimateWithResolution#

resolution integer (int64) required: Resolution of the output series specifying time interval between decimated points, in nanoseconds.

DerivativeSeries#

Calculates the rate of change between subsequent points.

input NumericSeries required
timeUnit TimeUnit: Time unit used to calculate the derivative. Defaults to seconds if not specified.
negativeValuesConfiguration NegativeValueConfiguration: Defines the strategy for handling negative output values. Defaults to allowNegativeValues if not specified.

DerivedSeries#

Represents a derived series within a compute graph. This is a series that is derived from a function within a module.

Set type to the name of one of the other fields, and set only that field.

function FunctionDerivedSeries

DetrendType#

The type of detrending to apply to the signal. When absent, will default to the detrend behavior for that function.

CONSTANT: Removes the mean value of the signal.
LINEAR: Removes a best-fit linear trend from the signal.
NO_OP: Does not perform any detrending.

Dimension#

A value derived per asset or run, injected as a tag.

tagKey StringConstant: The tag key to inject the value under. Defaults to the key derived by value.
value DimensionValue required: The value to inject.

DimensionsDataset#

Injects one tag per dimension onto every series of the underlying dataset, valued from the asset or run the series belongs to. This makes resource metadata usable as a grouping, filtering and aggregation dimension with selectTags, tag predicates and groupByTags. Group by the dimension together with ASSET_RID (or RUN_RID) to keep each resource separate, or by the dimension alone to pivot so that resources sharing a value fall into one group.

Dimensions are only supported on asset and run branches. Any other branch, such as a saved dataset or a derived series from withSeries, gets the reserved value NOMINAL_ABSENT for every dimension. To group derived series by a dimension, apply withDimensions to the withSeries input instead, so that the series they are computed from carry the dimension tags. A dimension tag replaces a series tag with the same key.

input Dataset required: The underlying dataset to tag.
dimensions array of Dimension: The dimensions to inject. Must not be empty, must have at most 10 entries, and every dimension must have a distinct tag key. A tag key must not be a ReservedTagKey, start with EVENT_LABEL_PREFIX:, or start with NOMINAL_.

DimensionValue#

Set type to the name of one of the other fields, and set only that field.

stringProperty StringPropertyDimension: The value of a string property on the asset or run.
identity IdentityDimension: The value of an asset’s primary key under one of its types.

Divide#

Divides left by right pointwise.

left NumericSeries required: Numerator.
right NumericSeries required: Denominator.

DoubleConstant#

A floating-point constant that can be a literal value, a named variable reference, or a named builtin constant.

Set type to the name of one of the other fields, and set only that field.

literal number: Create from a literal floating-point value.
variable string: Create from a named reference name.
builtin BuiltinConstant: Create from a named builtin constant; the server resolves the numeric value.

DriverSeries#

Uses the driver series timestamps as the alignment grid.

An empty JSON object, {}.

DriverSeries3d#

X
Y
Z

DropSeriesDataset#

Removes series from the base dataset by name: raw and derived series alike are no longer selectable under those names. Dropping a name absent from input has no effect. Combined with WithSeriesDataset, which can expose a series under another name, this renames a series: dropSeries(withSeries(input, [newName = select(oldName)]), [oldName]).

input Dataset required: The base dataset to drop series from.
names array of StringConstant: Names of the series to drop.

DurationAdd#

The sum of two durations.

left compute.api.Duration required: First duration.
right compute.api.Duration required: Second duration.

DurationConstant#

A duration constant that can be a literal value or a named variable reference.

Set type to the name of one of the other fields, and set only that field.

literal run.api.Duration: Create from a literal duration value.
variable string: Create from a named reference name.

DurationDays#

Constructs a duration from a number of days.

days IntegerConstant required: The number of days.

DurationFilterRanges#

Filters a list of ranges down to only those satisfying a threshold condition on the range’s duration.

input RangeSeries required
threshold DurationConstant required
operator ThresholdOperator required
unboundedBehavior UnboundedBehavior

DurationFilterRangesV2#

Filters ranges by duration using an authored Duration expression.

input RangeSeries required
duration compute.api.Duration required: The duration each input range is compared against.
unboundedBehavior UnboundedBehavior: How a range with an open start or end is measured. Defaults to treating the missing bound as the edge of the compute request window.
comparison ThresholdOperator required

DurationHours#

Constructs a duration from a number of hours.

hours IntegerConstant required: The number of hours.

DurationMilliseconds#

Constructs a duration from a number of milliseconds.

milliseconds IntegerConstant required: The number of milliseconds.

DurationMinutes#

Constructs a duration from a number of minutes.

minutes IntegerConstant required: The number of minutes.

DurationNanoseconds#

Constructs a duration from a number of nanoseconds. Limited to values within the 32-bit integer range (~2.1 billion ns, or ~2.1 seconds). For larger durations, use DurationSeconds or DurationMilliseconds.

nanoseconds IntegerConstant required: The number of nanoseconds.

DurationNegate#

Negates a duration (reverses its sign).

input compute.api.Duration required: The duration to negate.

DurationSeconds#

Constructs a duration from a number of seconds.

seconds IntegerConstant required: The number of seconds.

DurationSubtract#

The difference of two durations (left minus right).

left compute.api.Duration required: Duration to subtract from.
right compute.api.Duration required: Duration to subtract.

EarliestAnchorReference#

Selects the earliest anchor resolved from the input.

An empty JSON object, {}.

Empty#

An empty JSON object, {}.

Enum1dArraySelectTagsSeries#

Subsets the tag dimensions of the input enum 1d array series to the given tag keys. Preserves the input series’ cardinality: points that share the selected keys but differ on dropped keys remain distinct.

input Enum1dArraySeries required
tagKeys array of StringConstant

Enum1dArraySeries#

Set type to the name of one of the other fields, and set only that field.

channel ChannelSeries
raw Reference
derived DerivedSeries
selectEnum1dArray SelectSeries
extractFromStruct ExtractEnum1dArrayFromStructSeries
toStartOfInterval Enum1dArrayToStartOfIntervalSeries
timeShift Enum1dArrayTimeShiftSeries
filterByTag Enum1dArrayTagFilterSeries
selectTags Enum1dArraySelectTagsSeries

Enum1dArrayTagFilterSeries#

Applies a tag predicate to the enum 1d array series produced by input.

input Enum1dArraySeries required
predicate TagPredicate required

Enum1dArrayTimeShiftSeries#

input Enum1dArraySeries required
duration DurationConstant required

Enum1dArrayToStartOfIntervalSeries#

Snaps each point’s timestamp to the start of its enclosing fixed-size time bucket. Buckets are epoch-aligned: a point at timestamp t (nanoseconds) is rewritten to floor(t / window) * window. Values and tag groupings are preserved; aggregation of points that collide on the snapped timestamp is the responsibility of downstream nodes.

input Enum1dArraySeries required
window compute.api.Duration required

EnumAggregation#

Aggregates an input series using an enum aggregation operator. Rolling aggregation builders are not supported; only the group-by shape is accepted at resolution time. Min and Max require a categorical input series. Udf also accepts a numeric input, which reaches the function as Vec<f64>; a categorical input reaches it as Vec<String>.

input AggregationBuilder required: The aggregation shape to apply the operator over. Only the group-by shape is supported; rolling builders are rejected at resolution time. The Python enum_aggregation(...) sugar form is categorical-only. Numeric builders expose the UDF-only enum_udf(...) and rust_enum_udf(...) terminals instead.
operator EnumAggregationOperator required: The enum reduction applied to each group (min, max, or a user-defined function).

EnumAggregationFunction#

Aggregates duplicate timestamps on a single enum series. For aggregating across tag groupings, use EnumSeries.enumAggregation with aggregation.AggregationBuilder.groupBy.

MIN
MAX

EnumAggregationOperator#

Set type to the name of one of the other fields, and set only that field.

min Empty
max Empty
udf EnumAggregationUdf

EnumAggregationUdf#

A user-defined enum aggregation. The supported ABI is a Rust function with one vector input and a String return value. Numeric input series use Vec<f64> and enum input series use Vec<String>.

source UdfSource required

EnumBucket#

histogram map of string to integer (int64): The map of values within the bucket to their frequency.
firstPoint CompactEnumPoint required
lastPoint CompactEnumPoint: Will be empty if the bucket only has a single point.

EnumConstantResampleInterpolationConfiguration#

Fills all empty resample intervals with a single constant value.

If a resampled time interval (bucket) contains no data points from the input, it will be assigned the specified ‘constant’ as a default value. This applies to gaps at the start of the series, between data points, and at the end.

constant StringConstant required

EnumCountDuplicateSeries#

Counts the number of points along each timestamp in the input series.

inputs array of EnumSeries

EnumFilterOperator#

IS_IN: True when the value is contained in the set.
IS_NOT_IN: True when the value is not contained in the set.
MATCHES_ANY_REGEX: True when the value matches at least one pattern in the set. Values are RE2 regular expressions, anchored to the entire string.

EnumFilterRanges#

Produces a list of ranges for which the filter condition is satisfied.

input EnumSeries required
operator EnumFilterOperator required
values StringSetConstant required: The set of literal values (for IS_IN / IS_NOT_IN) or RE2 regex patterns (for MATCHES_ANY_REGEX) to match against.
persistenceWindowConfiguration PersistenceWindowConfiguration

EnumFilterTransformationSeries#

Outputs categorical input values that fall within the supplied ranges.

input EnumSeries required
filter RangeSeries required: Keeps only input points whose timestamp falls inside one of these ranges.

EnumHistogramBucket#

A bucket in an enum histogram representing all the counts for a specific enumerated value, across all input series.

value string required: The value of the enum as a string
countsByChannel map of VariableName to HistogramChannelCount

EnumHistogramNode#

inputs map of VariableName to EnumSeries

EnumHistogramPlot#

buckets array of EnumHistogramBucket

EnumMultivariateAggregation#

Aggregations for enum series in multivariate queries.

MODE

EnumNthPointDownsampleSeries#

Keeps every Nth row of the timestamp-ordered input, per series. No grid arithmetic, no interpolation, no boundary handling. Output count is ceil(input_count / n) per series. Caller is responsible for choosing n consistent with the input’s sample rate; output spacing is only uniform when the input is.

input EnumSeries required
n IntegerConstant required: Keep every Nth point starting from the earliest in each series. Must be >= 1.

EnumPlot#

timestamps array of Timestamp
values array of integer
categories array of string
nextPageToken PageToken: Continuation token for this paged result. Can only be returned if paging was requested. To retrieve the next page, repeat the same request with this value as pageToken and keep the pageSize sign unchanged. Returned only when this response contains exactly abs(pageSize) points; the next response may still be empty if this page ended exactly at the result-set boundary. Empty means this response was short and there are no further points in the requested direction.

EnumPoint#

timestamp Timestamp required
value string required

EnumResampleConfiguration#

interval DurationConstant required: Interval between resampled points
interpolation EnumResampleInterpolationConfiguration: Interpolation strategy to use (defaults to forward fill).

EnumResampleInterpolationConfiguration#

Set type to the name of one of the other fields, and set only that field.

forwardFillResampleInterpolationConfiguration ForwardFillResampleInterpolationConfiguration
constantResampleInterpolationConfiguration EnumConstantResampleInterpolationConfiguration

EnumResampleSeries#

Resamples the input series to a new resolution using interpolation. Outputs data for timestamps corresponding to the defined frequency. Based on interpolation strategy, determines range of timestamps to output data for and interpolates values where necessary.

input EnumSeries required
resampleConfiguration EnumResampleConfiguration required: The interval at which to resample the series and interpolation strategy

EnumResampleSeriesV2#

Resamples categorical points at the authored interval using forward-fill interpolation.

input EnumSeries required
interval compute.api.Duration required: Interval at which to resample.

EnumSelectTagsSeries#

Subsets the tag dimensions of the input enum series to the given tag keys. Preserves the input series’ cardinality: points that share the selected keys but differ on dropped keys remain distinct.

input EnumSeries required
tagKeys array of StringConstant

EnumSeries#

Set type to the name of one of the other fields, and set only that field.

aggregate AggregateEnumSeries deprecated: Deprecated: AggregateEnumSeries is deprecated. Use enumAggregation with aggregation.GroupByAggregationBuilder instead.
raw Reference
channel ChannelSeries
derived DerivedSeries
resample EnumResampleSeries
resampleV2 EnumResampleSeriesV2
nthPointDownsample EnumNthPointDownsampleSeries
timeRangeFilter EnumTimeRangeFilterSeries
timeShift EnumTimeShiftSeries
timeShiftV2 EnumTimeShiftSeriesV2
union EnumUnionSeries
filterTransformation EnumFilterTransformationSeries
valueMap ValueMapSeries
select1dArrayIndex SelectIndexFrom1dEnumArraySeries
extractFromStruct ExtractEnumFromStructSeries
eventAggregation EventsEnumSeries
enumAggregation EnumAggregation
selectEnum SelectSeries
selectCategoricalProperty SelectProperty
filterByTag EnumTagFilterSeries
selectTags EnumSelectTagsSeries
scalarUdf ScalarUdfSeries
toStartOfInterval EnumToStartOfIntervalSeries

EnumSeriesEqualityRanges#

Produces a list of ranges for which provided enum series are all equal (or are not all equal).

input array of EnumSeries
equalityOperator EqualityOperator required
persistenceWindowConfiguration PersistenceWindowConfiguration
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

EnumTagFilterSeries#

Applies a tag predicate to the enum series produced by input.

input EnumSeries required
predicate TagPredicate required

EnumTimeRangeFilterSeries#

Filters the series to points within the specified time range.

input EnumSeries required
startTime TimestampConstant: Represents the start time (inclusive) of the time range.
endTime TimestampConstant: Represents the end time (inclusive) of the time range.

EnumTimeShiftSeries#

input EnumSeries required
duration DurationConstant required

EnumTimeShiftSeriesV2#

Shifts the timestamps in the input series by the resolved duration.

input EnumSeries required
duration compute.api.Duration required: Duration to shift the input series by.

EnumToNumericSeries#

Maps an enumerated series to a numeric series by mapping each string value to a double.

input EnumSeries required
mapping map of string to DoubleConstant: The mapping from enum values to doubles.
defaultValue DoubleConstant: The value to use for enum values not present in the mapping. If not specified, points with unmapped enum values will be dropped.

EnumToStartOfIntervalSeries#

input EnumSeries required
window compute.api.Duration required

EnumUnionOperation#

THROW: Throws an exception if two series contain points with a duplicate timestamp.
MAX: Aggregates duplicates by String MAX.
MIN: Aggregates duplicates by String MIN.

EnumUnionSeries#

Combines multiple enum series together and outputs a single series. The strategy to merge input values with the same timestamp together is specified in the operation field.

input array of EnumSeries
operation EnumUnionOperation required: The strategy to merge points with duplicate timestamps.

EqualityOperator#

ALL_EQUAL
NOT_ALL_EQUAL

EqualToSeries#

Compares two numeric series point-wise, producing true where left == right and false otherwise.

left NumericSeries required
right NumericSeries required

ErrorResult#

errorType string required
code integer required

EventAnchor#

Uses the start timestamp of the event at the configured zero-based index from this query whose event time lies in the anchor window [start, end), or [start, +infinity) when the end is absent. The window is the run bounds for a run anchor. Shares the wire shape of numeric.EventTimeShift, which tagByIntervals time shifts use; a separate type so dataset anchors evolve independently.

query ComputeEventQuery required
index IntegerConstant: Zero-based event index after sorting. Defaults to zero and must be non-negative.
sortOrder EventSortOrder: Ordering by event start timestamp before selecting the index. Defaults to ASC.

EventDispositionStatus#

PENDING_REVIEW
CLOSED_IGNORED
CLOSED_REQUIRES_FURTHER_ACTION
NO_DISPOSITION

EventIntervalSource#

Intervals come from an event query. Each event defines [timestamp, timestamp+duration).

query ComputeEventQuery required
tagBy array of EventTagSource: Non-empty ordered list of tag sources. Each contributes a tag key-value pair per event interval.

EventLabelTag#

label StringConstant required

EventLevelTag#

An empty JSON object, {}.

EventPropertyTag#

propertyName StringConstant required

EventsEnumSeries#

Queries events and emits an enum series where each data point corresponds to an event and the enum value is derived from a single event field (property, label, etc).

query ComputeEventQuery required
valueSource EventsEnumValueSource required

EventsEnumValueSource#

Set type to the name of one of the other fields, and set only that field.

property StringConstant: Name of the event property whose value should be used as the enum value.
level Empty: The event level (INFO, FLAG, VIOLATION, SUCCESS). This field corresponds to the event type.

EventSortOrder#

ASC
DESC

EventsSearchRanges#

Produces a range series with a range for each event matching the query. Throws if there are more than 1,000 results.

query ComputeEventQuery required

EventTagSource#

A dimension contributing a tag key-value pair per event interval.

Set type to the name of one of the other fields, and set only that field.

intervalIndex IntervalIndexTag
level EventLevelTag
property EventPropertyTag
label EventLabelTag
rangeStartTimestamp IntervalStartTimestampTag

EventTimeFilter#

timestamp TimestampConstant required
timestampCondition EventTimeFilterCondition required

EventTimeFilterCondition#

START_TIME_BEFORE_INCLUSIVE
START_TIME_BEFORE_EXCLUSIVE
START_TIME_AFTER_INCLUSIVE
START_TIME_AFTER_EXCLUSIVE
END_TIME_BEFORE_INCLUSIVE
END_TIME_BEFORE_EXCLUSIVE
END_TIME_AFTER_INCLUSIVE
END_TIME_AFTER_EXCLUSIVE

EventTimeShift#

Shift by the start timestamp of the event at the configured zero-based index from this query whose event time lies in the anchor window [start, end), or [start, +infinity) when the end is absent. The window is the tagging interval for tagByIntervals and the run bounds for a run anchor. The query may differ from the query that produced a tagging interval.

query ComputeEventQuery required
index IntegerConstant: Zero-based event index after sorting. Defaults to zero and must be non-negative.
sortOrder EventSortOrder: Ordering by event start timestamp before selecting the index. Defaults to ASC.

EventType#

INFO
FLAG
ERROR
SUCCESS

EveryNthPerSeriesSampling#

Returns every Nth timestamp-ordered point from each distinct series.

n IntegerConstant required: Keep every Nth point starting from the earliest in each series. Must be > 1.

ExcludeNegativeValues#

An empty JSON object, {}.

ExponentialCurve#

An empty JSON object, {}.

ExponentialResultDetails#

y = a * e^(b * x)

a number required
b number required

ExtractEnum1dArrayFromStructSeries#

input StructSeries required
fieldPath array of StringConstant

ExtractEnumFromStructSeries#

input StructSeries required
fieldPath array of StringConstant: Path segments that are joined to locate a nested enum field. Use dots to traverse objects (e.g. a.b.c) and 0-based brackets to index into arrays (e.g. items[0].value). A bare numeric name like 0 refers to an object key named “0”; use [0] to index into an array. Rows where the path is invalid or resolves to a nested object rather than a scalar are filtered out.

ExtractNumeric1dArrayFromStructSeries#

input StructSeries required
fieldPath array of StringConstant

ExtractNumericFromStructSeries#

input StructSeries required
fieldPath array of StringConstant: Path segments that are joined to locate a nested numeric field. Use dots to traverse objects (e.g. a.b.c) and 0-based brackets to index into arrays (e.g. items[0].value). A bare numeric name like 0 refers to an object key named “0”; use [0] to index into an array. Rows where the path is invalid or the value cannot be converted to the specified type are filtered out.
dataType NumericDataType: The type the numeric data will be cast to, defaults to FLOAT64

ExtractStructFromStructSeries#

input StructSeries required
fieldPath array of StringConstant: Path segments that are joined to locate a nested struct field. Use dots to traverse objects (e.g. a.b.c) and 0-based brackets to index into arrays (e.g. items[0].value). A bare numeric name like 0 refers to an object key named “0”; use [0] to index into an array. Returns an empty struct when the path does not resolve to a valid JSON object.

ExtremaRangesV2#

Produces ranges for the selected extrema type.

input NumericSeries required
minimumProminence DoubleConstant: The minimum topographic prominence for an extrema to be returned.
selection PeakType required

Fft#

Returns the single sided amplitude spectrum of the input series.

input NumericSeries required
window FftWindow: Window function applied to the input series. Defaults to RECT is not specified.
summarizationStrategy FrequencySummarizationStrategy: Strategy to downsample the output frequency spectrum.
nfft integer: Length of the FFT. When nfft < input length, the input is truncated to the first nfft samples before windowing and transforming. When nfft > input length, the windowed input is zero-padded up to nfft before transforming. When nfft is unset, defaults to the input length, producing the same result as setting nfft to the input length. The output contains nfft/2 + 1 single-sided frequency bins with spacing fs/nfft. For best performance, prefer a power of 2 (e.g. 512, 1024, 2048).

FftWindow#

BLACKMAN
HANN
HAMMING
RECT

FillLimit#

How far forward in time a sample’s value may be carried.

Set type to the name of one of the other fields, and set only that field.

duration DurationConstant

FillStrategy#

Default strategy for filling gaps when aligning multi-input series.

Set type to the name of one of the other fields, and set only that field.

forwardFill ForwardFill

FilterByExpressionSeries#

Keeps points from the base series where the boolean expression evaluates to true. Only outputs timestamps where all input series have an entry for that timestamp, or a value can be filled using the interpolation configuration.

base string required: The variable name for the series that will be returned by this filter. The expression can filter points conditional on multiple series. The base defines which series’ points will be returned after applying all the filters.
inputs map of LocalVariableName to NumericSeries: A map containing all the numeric series present in the expression.
expression string required: An expression that evaluates to a boolean. For example - a > 5.
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

FilteredDataset#

Applies a conjunctive tag predicate to the series produced by the inner dataset. Resolved by threading tag filters into channel specs and folding against known tag assignments; does not add a resolved numeric node.

input Dataset required: The underlying dataset to filter.
predicate TagPredicate required: Conjunctive tag predicate used to keep matching series.

FirstPointMatchingCondition#

The first point in the output range will be the first point where the condition evaluates to true.

An empty JSON object, {}.

FloorDivide#

Divides left by right pointwise and rounds down to the nearest integer.

left NumericSeries required: Numerator.
right NumericSeries required: Denominator.

ForwardFill#

At each grid point, carries the most recent prior sample value forward. The value is held only up to the configured limit; grid points with no prior sample within that window are left empty.

limit FillLimit required

ForwardFillInterpolation#

interpolationRadius DurationConstant required

ForwardFillResampleInterpolationConfiguration#

Forward fill interpolation for resampling. Rounds the earliest and latest timestamp up to the nearest interval multiple and generates timestamps at the given interval between the new earliest and latest timestamps. For every timestamp in the resampled timestamps, takes the last known value before the timestamp as the value.

An empty JSON object, {}.

FrequencyBucket#

A single bucket in a bucketed frequency domain plot.

frequencyStart number required: The start frequency of this bucket (inclusive).
frequencyEnd number required: The end frequency of this bucket (exclusive).
frequencyCenter number required: The center frequency of this bucket.
minAmplitude number required: The minimum amplitude value in this bucket.
maxAmplitude number required: The maximum amplitude value in this bucket.
meanAmplitude number required: The mean amplitude value in this bucket.
count integer (int64) required: The number of original frequency points that were aggregated into this bucket.
firstPoint FrequencyPoint required: The first point (lowest frequency) in this bucket.
lastPoint FrequencyPoint: The last point (highest frequency) in this bucket. Will be empty if the bucket only has a single point.

FrequencyDecimateWithBuckets#

buckets integer required: Target number of frequency bins/buckets.
minFrequency number: Optional minimum frequency bound. When set, frequencies below this value are excluded before bucketing, increasing resolution within the requested range. This bound is inclusive. The bound uses the same unit as the transform’s output frequency.
maxFrequency number: Optional maximum frequency bound. When set, frequencies above this value are excluded before bucketing, increasing resolution within the requested range. This bound is exclusive. The bound uses the same unit as the transform’s output frequency.

FrequencyDomain#

Set type to the name of one of the other fields, and set only that field.

fft Fft
periodogram Periodogram

FrequencyDomainPlot#

frequencies array of number
amplitudes array of number

FrequencyDomainPlotComplex#

frequencies array of number
real array of number
imaginary array of number

FrequencyDomainPlotMagnitudeAndPhase#

frequencies array of number
magnitudes array of number
phases array of number

FrequencyDomainPlotV2#

Set type to the name of one of the other fields, and set only that field.

real FrequencyDomainPlot
complex FrequencyDomainPlotComplex
magnitudeAndPhase FrequencyDomainPlotMagnitudeAndPhase

FrequencyDomainV2#

Set type to the name of one of the other fields, and set only that field.

fft Fft
psd Psd
cpsd Cpsd
nyquist Nyquist
bode Bode

FrequencyPoint#

A single point in a frequency domain plot.

frequency number required
amplitude number required

FrequencySummarizationStrategy#

Set type to the name of one of the other fields, and set only that field.

buckets FrequencyDecimateWithBuckets

FunctionDerivedSeries#

A reference to a derived series resulting from applying a function to a set of arguments.

moduleName StringConstant required
functionName StringConstant required
versionReference ModuleVersionReference required
functionArgs map of FunctionParameterName to FunctionParameterValue: Map of function input names to their values. The function inputs must match the function’s parameter names and types. An input must be specified for each of the referenced function’s parameters.

FunctionParameterValue#

Set type to the name of one of the other fields, and set only that field.

variable string
value VariableValue

FunctionVariables#

variables map of VariableName to VariableValue
subFunctionVariables map of FunctionReference to FunctionVariables

GreaterThanOrEqualToSeries#

Compares two numeric series point-wise, producing true where left >= right and false otherwise.

left NumericSeries required
right NumericSeries required

GreaterThanSeries#

Compares two numeric series point-wise, producing true where left > right and false otherwise.

left NumericSeries required
right NumericSeries required

GroupByAggregationBuilder#

Groups points by a tag selection and a timestamp policy and applies the operator.

input Series required: The series to group and aggregate.
tags GroupByTags required: Selects which tag keys form the grouping key: GroupByTags.Current() keeps the input series’ currently selected tags, or GroupByTags.Subset([...]) restricts the grouping to a subset of them.
timestamp GroupByTimestamp required: Selects how timestamps participate in the grouping key: GroupByTimestamp.Current() groups on existing timestamps, and GroupByTimestamp.Every(duration) buckets points into fixed-size, epoch-aligned tumbling windows.
alignment GroupByAlignment: Optionally aligns source tag groups that map to the same output tag group before aggregation. Alignment is currently supported only with GroupByTimestamp.Current() and built-in numeric aggregation operators.

GroupByAlignment#

Selects how source tag groups are aligned before aggregation.

Set type to the name of one of the other fields, and set only that field.

contextDefaults Empty: Use the compute context’s defaultAlignmentStrategy and defaultFillStrategy.

GroupBySubsetTags#

tags array of StringConstant: Tag keys to group by.

GroupByTags#

Selects which tag keys form the grouping key.

Set type to the name of one of the other fields, and set only that field.

current Empty: Group by the tags currently selected on the input series.
subset GroupBySubsetTags: Group by exactly the listed tag keys. Must be a (non-strict) subset of the input series’s currently selected tags.

GroupByTimestamp#

Selects how timestamps participate in the grouping key.

Set type to the name of one of the other fields, and set only that field.

current Empty: Use the current timestamps.
every compute.api.Duration: Bucket timestamps into a tumbling fixed-size, epoch-aligned window. A point at timestamp t (nanoseconds) falls into the bucket [floor(t / window) * window, +window), and one aggregated point is emitted per bucket per tag grouping at the bucket start timestamp.

GroupedComputeNodeResponse#

grouping Grouping required
response ComputeNodeResponse required

GroupedComputeNodeResponses#

Contains a ComputeNodeResponse for each applicable grouping along with metadata describing the grouping. All the contained ComputeNodeResponses are guaranteed to be of the same type.

responses array of GroupedComputeNodeResponse

Grouping#

Set type to the name of one of the other fields, and set only that field.

tagsWithValues map of string to string: A grouping identified by a specific instantiation of tag keys and values.

GroupWithinRangesSeriesV2#

Aggregates the input series within each supplied range and emits one point at the range start. Empty ranges emit no point, and ranges without a start use the compute request start.

input NumericSeries required
ranges RangeSeries required: One output point is emitted per range, at the range’s start.
aggregation NumericAggregationOperator required

HashSubsampleSampling#

Keeps approximately 1/N of raw points via a stateless hash of each point’s timestamp.

n IntegerConstant required: Keep approximately one out of every N points. Must be > 1.

HighPassConfiguration#

cutoffFrequency DoubleConstant required

Histogram#

Set type to the name of one of the other fields, and set only that field.

numeric NumericHistogramNode
enum EnumHistogramNode

HistogramChannelCount#

The count of the value in the bucket for the specific channel.

count integer (int64) required

IdentityDimension#

Injects the value of a type’s primary key, under a tag key equal to the primary key’s property name by default. The primary key is read when the query runs, so it follows changes to the type. Only asset branches of an asset with the type have a value; every other branch, including run branches, gets the reserved value NOMINAL_ABSENT.

typeRid StringConstant required: Resource identifier of the type whose primary key identifies the asset.

IncompatibleUnitOperation#

Attempted an incompatible operation of units.

operation UnitOperation required
units array of string

IntegerConstant#

An integer constant that can be a literal value or a named variable reference.

Set type to the name of one of the other fields, and set only that field.

literal integer: Create from a literal integer value.
variable string: Create from a named reference name.

IntegralSeries#

Calculates the running sum of the area underneath a series using the trapezoidal rule.

input NumericSeries required
startTimestamp TimestampConstant required
timeUnit TimeUnit: Time unit used to calculate the integral. Defaults to seconds if not specified.

InterpolationConfiguration#

Set type to the name of one of the other fields, and set only that field.

forwardFillInterpolation ForwardFillInterpolation

IntersectRanges#

Produces the intersection of the input ranges: the ranges of time covered by every input. For best performance, pass inputs in the order in which they should be evaluated, listing the ranges that are cheapest to compute first.

inputs array of RangeSeries

IntervalAnchor#

Selects which boundary of an interval to use as its anchor.

Set type to the name of one of the other fields, and set only that field.

intervalStart IntervalStartTimeShift: Uses the interval’s start timestamp.
intervalEnd IntervalEndTimeShift: Uses the interval’s end timestamp.
eventInInterval EventAnchor
rangeInInterval RangesAnchor

IntervalEndTimeShift#

Shift by each tagging interval’s resolved end; resolution throws if end is missing.

An empty JSON object, {}.

IntervalIndexTag#

An empty JSON object, {}.

IntervalSource#

Where the intervals come from and how each is labeled.

Set type to the name of one of the other fields, and set only that field.

events EventIntervalSource
ranges RangeIntervalSource

IntervalStartTimeShift#

Shift by each tagging interval’s resolved start; resolution throws if start is missing.

An empty JSON object, {}.

IntervalStartTimestampTag#

An empty JSON object, {}.

LabelsFilter#

operator SetOperator required
labels array of string

LessThanOrEqualToSeries#

Compares two numeric series point-wise, producing true where left <= right and false otherwise.

left NumericSeries required
right NumericSeries required

LessThanSeries#

Compares two numeric series point-wise, producing true where left < right and false otherwise.

left NumericSeries required
right NumericSeries required

LiteralRange#

A literal range of values.

startTimestamp TimestampConstant
endTimestamp TimestampConstant
tags map of string to string

LiteralRanges#

A list of literal ranges.

literalRanges array of LiteralRange

Ln#

Computes the natural logarithm (base e) of each value.

x NumericSeries required: Input numeric series (positive values).

Log10#

Computes the base-10 logarithm of each value.

x NumericSeries required: Input numeric series (positive values).

LogArgFilterOperator#

Matches against a single log arg.

key StringConstant required
operator LogRegexFilterOperator required

LogarithmicCurve#

An empty JSON object, {}.

LogarithmicResultDetails#

y = a * ln(x) + b

a number required
b number required

LogExactMatchCaseInsensitiveFilter#

Filters points such that the log message in each point contains an exact case-insensitive match of the provided token.

token string required

LogFilterOperator#

Set type to the name of one of the other fields, and set only that field.

regexFilter LogRegexFilterOperator
exactMatchCaseInsensitiveFilter LogExactMatchCaseInsensitiveFilter
arg LogArgFilterOperator
and array of LogFilterOperator
or array of LogFilterOperator
not LogFilterOperator

LogFilterSeries#

Outputs only values of the log plot that satisfy the filter.

input LogSeries required
operator LogFilterOperator required

LogPoint#

timestamp Timestamp required
value LogValue required

LogRegexFilterOperator#

Matches a text value against the given re2 regular expression. Regular expression syntax: https://github.com/google/re2/wiki/Syntax.

regex string required

LogSelectTagsSeries#

Subsets the tag dimensions of the input log series to the given tag keys. Preserves the input series’ cardinality: points that share the selected keys but differ on dropped keys remain distinct.

input LogSeries required
tagKeys array of StringConstant

LogSeries#

Set type to the name of one of the other fields, and set only that field.

raw Reference
channel ChannelSeries
derived DerivedSeries
selectLogs SelectSeries
union LogUnionSeries
filter LogFilterSeries
timeShift LogTimeShiftSeries
toStartOfInterval LogToStartOfIntervalSeries
filterByTag LogTagFilterSeries
selectTags LogSelectTagsSeries

LogTagFilterSeries#

Applies a tag predicate to the log series produced by input.

input LogSeries required
predicate TagPredicate required

LogTimeShiftSeries#

input LogSeries required
duration DurationConstant required

LogToStartOfIntervalSeries#

input LogSeries required
window compute.api.Duration required

LogUnionOperation#

KEEP_ALL: Keeps all points if multiple series contain points with a duplicate timestamp.

LogUnionSeries#

Combines multiple log series together and outputs a single series. The strategy to merge input values with the same timestamp together is specified in the operation field.

input array of LogSeries
operation LogUnionOperation required: The strategy to merge points with duplicate timestamps.

LogValue#

message string required
args map of string to string
id string (UUID) required

LowPassConfiguration#

cutoffFrequency DoubleConstant required

LttbStrategy#

maxPointsPerGroup integer required: Maximum number of selected points to return for each group. Must be positive and no greater than 10,000.

MagnitudeScaling#

LINEAR: Linear scaling of the magnitude.
MAGNITUDE_DB_10: Scales the magnitude via 10 * log10(magnitude).
MAGNITUDE_DB_20: Scales the magnitude via 20 * log10(magnitude).

MathConstant#

Math constants referenceable by name.

PI: The circle constant π (dimensionless).
TAU: The circle constant τ = 2π (dimensionless).
E: Euler’s number e (dimensionless).

Maximum#

The maximum value of points inside the time window.

An empty JSON object, {}.

MaxSeries#

For every timestamp specified in the input series, outputs a value that is the maximum for that timestamp across all input series. Only outputs timestamps where all input series have an entry for that timestamp, subject to interpolation.

inputs array of NumericSeries: Input series across which to take the pointwise maximum
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

MeanSeries#

For every timestamp specified in the input series, outputs a value that is the mean for that timestamp across all input series. Only outputs timestamps where all input series have an entry for that timestamp, subject to interpolation.

inputs array of NumericSeries: Input series across which to take the pointwise mean
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

Minimum#

The minimum value of points inside the time window.

An empty JSON object, {}.

MinMaxThresholdOperator#

OUTSIDE_BOUNDS
OUTSIDE_OR_EQUAL_TO_BOUNDS
INSIDE_BOUNDS
INSIDE_OR_EQUAL_TO_BOUNDS

MinMaxThresholdRanges#

Produces a list of ranges for which the threshold condition is satisfied.

input NumericSeries required
lowerBound DoubleConstant required
upperBound DoubleConstant required
operator MinMaxThresholdOperator required
persistenceWindowConfiguration PersistenceWindowConfiguration

MinSeries#

For every timestamp specified in the input series, outputs a value that is the minimum for that timestamp across all input series. Only outputs timestamps where all input series have an entry for that timestamp, subject to interpolation.

inputs array of NumericSeries: Input series across which to take the pointwise minimum
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

ModuleVersionReference#

Set type to the name of one of the other fields, and set only that field.

pinned PinnedModuleVersionReference

Modulo#

Computes the remainder of left divided by right pointwise.

left NumericSeries required: Dividend.
right NumericSeries required: Divisor.

Multiply#

Multiplies two numeric series pointwise.

left NumericSeries required: Left-hand operand.
right NumericSeries required: Right-hand operand.

MultivariateBucket#

timestamp Timestamp required
values array of Value: Values for each input series in the bucket.

MultivariateEnumInput#

series EnumSeries required
aggregations array of EnumMultivariateAggregation

MultivariateInput#

Set type to the name of one of the other fields, and set only that field.

numeric MultivariateNumericInput
enum MultivariateEnumInput

MultivariateNumericInput#

series NumericSeries required
aggregations array of NumericMultivariateAggregation

MultivariateUnitResult#

units array of UnitResult

NamedSeries#

name StringConstant required: Name of the new derived series.
definition Series required: Compute expression defining the series values.

Negate#

Negates each value in the series pointwise (multiplies by -1).

x NumericSeries required: Input numeric series.

NegativeValueConfiguration#

Set type to the name of one of the other fields, and set only that field.

allowNegativeValues AllowNegativeValues: Include negative output values
setNegativeValuesToZero SetNegativeValuesToZero: Set negative output values to zero
excludeNegativeValues ExcludeNegativeValues: Exclude points with negative output values

NotEqualToSeries#

Compares two numeric series point-wise, producing true where left != right and false otherwise.

left NumericSeries required
right NumericSeries required

NotRanges#

The not ranges node will invert the ranges, filling the negative space in time.

input RangeSeries required

NthRange#

Select the Nth range (0-indexed) from a range series.

input RangeSeries required
index IntegerConstant required
sortOrder RangeSortOrder required

Numeric1dArraySelectTagsSeries#

Subsets the tag dimensions of the input numeric 1d array series to the given tag keys. Preserves the input series’ cardinality: points that share the selected keys but differ on dropped keys remain distinct.

input Numeric1dArraySeries required
tagKeys array of StringConstant

Numeric1dArraySeries#

Set type to the name of one of the other fields, and set only that field.

channel ChannelSeries
raw Reference
derived DerivedSeries
selectNumeric1dArray SelectSeries
extractFromStruct ExtractNumeric1dArrayFromStructSeries
toStartOfInterval Numeric1dArrayToStartOfIntervalSeries
timeShift Numeric1dArrayTimeShiftSeries
filterByTag Numeric1dArrayTagFilterSeries
selectTags Numeric1dArraySelectTagsSeries

Numeric1dArrayTagFilterSeries#

Applies a tag predicate to the numeric 1d array series produced by input.

input Numeric1dArraySeries required
predicate TagPredicate required

Numeric1dArrayTimeShiftSeries#

input Numeric1dArraySeries required
duration DurationConstant required

Numeric1dArrayToStartOfIntervalSeries#

input Numeric1dArraySeries required
window compute.api.Duration required

NumericAggregation#

Aggregates an input series using the given operator. The shape of the aggregation (rolling time window, rolling point-count window, or group by) is selected by the aggregation builder passed as input. The operator must be compatible with that builder’s series, producing a numeric output: count reduces a series of any type, whereas SUM requires a numeric series.

input AggregationBuilder required: The aggregation shape to apply the operator over: a rolling time window, a rolling point-count window, or a group-by.
operator NumericAggregationOperator required: The reduction applied to each window or group (e.g. sum, mean, percentile). Must be compatible with the input series, producing a numeric output: Count works on any input series, whereas reductions like Sum require a numeric series.

NumericAggregationFunction#

DEPRECATED. Use NumericAggregationOperator instead.

SUM
MEAN
MIN
MAX
COUNT
STANDARD_DEVIATION
ROOT_MEAN_SQUARE

NumericAggregationOperator#

Unified numeric aggregation used across rolling windows, plain aggregates, range aggregates, union merges, and per-bucket summarization.

Set type to the name of one of the other fields, and set only that field.

sum Summation
average Average
min Minimum
max Maximum
count Count
standardDeviation StandardDeviation
rootMeanSquare RootMeanSquare
percentile Percentile
udf NumericAggregationUdf

NumericAggregationUdf#

A user-defined numeric aggregation. The aggregation builder controls grouping/windowing. Supported Rust entrypoints accept grouped value vectors, optionally with leading sample timestamps, and return either one numeric value or timestamped numeric point(s), e.g. fn rms(values: Vec<f64>) -> f64, fn energy(timestamp_ns: Vec<i64>, values: Vec<f64>) -> f64, fn first(timestamp_ns: Vec<i64>, values: Vec<f64>) -> (i64, f64), or fn passthrough(timestamp_ns: Vec<i64>, values: Vec<f64>) -> Vec<(i64, f64)>. The value vector’s element type follows the aggregated series: Vec<f64> for a numeric series, and Vec<String> for a categorical or struct series, where each struct row arrives as a JSON object. A UDF over a combined struct series may instead take one vector per named struct field it needs, e.g. fn mean_product(channel0: Vec<f64>, channel1: Vec<f64>) -> f64. Parameter names select the fields, so a subset of the struct’s fields in any order is accepted.

source UdfSource required

NumericApproximateFilterSeries#

Outputs the values of the numeric plot value that are approximately equal to the threshold value.

input NumericSeries required
threshold DoubleConstant required
tolerance DoubleConstant required
operator ApproximateThresholdOperator required

NumericBucket#

min number
max number
mean number
count integer (int64)
variance number: The population variance of the bucket. If the bucket has only one value, this will be 0.
firstPoint NumericPoint
lastPoint NumericPoint: Will be empty if the bucket only has a single point.
aggregations map of string to number: Additional per-bucket aggregations that have no dedicated field above, keyed by the output column name requested via numericAggregations or numericOutputFieldsV2 (e.g. sum, standardDeviation, rootMeanSquare, p50). Empty when no additional aggregations were requested.

NumericConstantResampleInterpolationConfiguration#

Fills all empty resample intervals with a single constant value.

constant DoubleConstant required

NumericDataType#

Supported numeric data types for value casting.

INT64
FLOAT64
UINT64

NumericEqualityRangesV2#

Produces ranges where all numeric inputs are equal or are not all equal.

inputs array of NumericSeries: The series compared against each other. At least two are required.
tolerance DoubleConstant: Maximum spread between the inputs still counted as equal. Defaults to zero, i.e. exact equality.
minPoints IntegerConstant: Minimum number of matching points. Defaults to 1.
minDuration compute.api.Duration: Minimum matching duration. Defaults to one nanosecond.
rangeStart RangeStart: Selects whether the output begins at the first match or after persistence is satisfied.
equalityOperator EqualityOperator required

NumericFilterTransformationSeries#

Outputs the input points whose timestamps fall within the supplied ranges.

input NumericSeries required
filter RangeSeries required: Keeps input points whose timestamps fall inside any of these ranges.

NumericFilterWhereSeries#

Outputs numeric input values where the provided Boolean expression evaluates to true.

input NumericSeries required
condition BooleanSeries required: Condition evaluated point-wise at each input timestamp. A point is only kept where the condition is true.

NumericHistogramBucket#

A bucket in a numeric histogram representing a range of values, and the counts of values in that range across all input series.

lowerBound number required: The lower bound of the bucket, inclusive
upperBound number required: The upper bound of the bucket, exclusive
countsByChannel map of VariableName to HistogramChannelCount
buckets map of VariableName to HistogramChannelCount deprecated: Deprecated: Use countsByChannel instead

NumericHistogramBucketStrategy#

Set type to the name of one of the other fields, and set only that field.

bucketCount IntegerConstant: The number of buckets to use in the histogram. Width is automatically calculated from the range of the input series, with the lower and upper bounds of the histogram being multiples of the width.
bucketWidthAndOffset NumericHistogramBucketWidthAndOffset

NumericHistogramBucketWidthAndOffset#

width DoubleConstant required: The width of each bucket. If offset is not specified, the left and right bounds of each bucket are multiples of the width. If offset is specified, the left and right bounds of each bucket are shifted right by the offset value.
offset DoubleConstant: The rightward shift to apply to the left and right bounds of each bucket. If not specified, the offset will be 0. The offset must be non-negative.

NumericHistogramNode#

inputs map of VariableName to NumericSeries
bucketStrategy NumericHistogramBucketStrategy

NumericHistogramPlot#

buckets array of NumericHistogramBucket

NumericIfElseSeries#

Selects between two numeric series point-wise based on a boolean condition series. For every aligned timestamp, emits the value from trueValue where condition is true, and the value from falseValue otherwise.

condition BooleanSeries required: Boolean series evaluated at each point to choose between the trueValue and falseValue branches.
trueValue NumericSeries required: Value emitted at points where condition is true.
falseValue NumericSeries required: Value emitted at points where condition is false.

NumericLargestTriangleThreeBucketsSeries#

Applies Largest-Triangle-Three-Buckets to a numeric time series for visualization-oriented downsampling. The algorithm preserves the first and last points, partitions the remaining input points based on the requested output count, and selects points that preserve the visual shape of the series. Results are selected input timestamp/value points, not fixed-width bucket aggregate rows.

input NumericSeries required
maxPointsToReturn IntegerConstant required: Maximum number of selected points to return. Must be positive and no greater than 10,000.

NumericMultivariateAggregation#

MEAN
MIN
MAX
FIRST
LAST

NumericNanFilter#

Removes points where the value is NaN from the input series.

input NumericSeries required: Input numeric series.

NumericNthPointDownsampleSeries#

input NumericSeries required
n IntegerConstant required: Keep every Nth point starting from the earliest in each series. Must be >= 1.

NumericOutputField#

MIN
MAX
MEAN
COUNT
VARIANCE
FIRST_POINT
LAST_POINT

NumericOutputFieldV2#

Numeric bucket output field requested by the caller.

Set type to the name of one of the other fields, and set only that field.

field NumericOutputFieldV2Name
percentile NumericOutputPercentile

NumericOutputFieldV2Name#

MIN
MAX
MEAN
COUNT
VARIANCE
FIRST_POINT
LAST_POINT
SUM
STANDARD_DEVIATION
ROOT_MEAN_SQUARE

NumericOutputPercentile#

percentile DoubleConstant required: Percentile to compute, in range [0, 1].
outputName string: Optional output column name. Defaults to pXX, e.g. p95 for 0.95.

NumericPlot#

timestamps array of Timestamp
values array of number
nextPageToken PageToken: Continuation token for this paged result. Can only be returned if paging was requested. To retrieve the next page, repeat the same request with this value as pageToken and keep the pageSize sign unchanged. Returned only when this response contains exactly abs(pageSize) points; the next response may still be empty if this page ended exactly at the result-set boundary. Empty means this response was short and there are no further points in the requested direction.

NumericPoint#

timestamp Timestamp required
value number required

NumericPropertyPredicate#

Matches resources that have a numeric value for the named property and whose value satisfies the operator. NEQ excludes resources that lack the property or store a string value under the same name.

name string required
value number required
operator PropertyComparisonOperator required

NumericPropertyRangeOperator#

BETWEEN
NOT_BETWEEN

NumericPropertyRangePredicate#

Inclusive range on a numeric property. At least one bound is required. BETWEEN: min <= x <= max. An absent bound is unbounded on that side. NOT_BETWEEN: x < min OR x > max. An absent bound drops that side. Both operators require a numeric value for name. Missing properties and string values never match. Bounds are closed [min, max] to match the number chip (is between / is not between), not half-open.

name string required
min number
max number
operator NumericPropertyRangeOperator required

NumericResampleConfiguration#

interval DurationConstant required: Interval between resampled points
interpolation NumericResampleInterpolationConfiguration: Interpolation strategy to use (defaults to forward fill).

NumericResampleInterpolationConfiguration#

Set type to the name of one of the other fields, and set only that field.

forwardFillResampleInterpolationConfiguration ForwardFillResampleInterpolationConfiguration
constantResampleInterpolationConfiguration NumericConstantResampleInterpolationConfiguration

NumericResampleSeries#

input NumericSeries required
resampleConfiguration NumericResampleConfiguration required: The interpolation strategy and interval at which to resample the series

NumericResampleSeriesV2#

Resamples the input series at the authored interval using forward-fill.

input NumericSeries required
interval compute.api.Duration required: Interval at which to resample.

NumericSelectTagsSeries#

Subsets the tag dimensions of the input numeric series to the given tag keys. Preserves the input series’ cardinality: points that share the selected keys but differ on dropped keys remain distinct.

input NumericSeries required
tagKeys array of StringConstant

NumericSeries#

Set type to the name of one of the other fields, and set only that field.

abs Abs
negate Negate
cos Cos
sin Sin
tan Tan
acos Acos
asin Asin
ln Ln
log10 Log10
sqrt Sqrt
add Add
subtract Subtract
multiply Multiply
divide Divide
floorDivide FloorDivide
power Power
modulo Modulo
atan2 Atan2
max MaxSeries
mean MeanSeries
min MinSeries
sum SumSeries
union NumericUnionSeries
product ProductSeries
constant ConstantNumericSeries
selectNewestPoints SelectNewestPointsSeries
selectNumeric SelectSeries
selectNumericProperty SelectProperty
selectOldestPoints SelectOldestPointsSeries
bitAnd BitAnd
bitOr BitOr
bitXor BitXor
bitShiftRight BitShiftRight
bitShiftLeft BitShiftLeft
bitTest BitTest
ifElse NumericIfElseSeries
numericAggregation NumericAggregation
countDuplicate EnumCountDuplicateSeries
cumulativeSum CumulativeSumSeries
derivative DerivativeSeries
integral IntegralSeries
zScore ZscoreSeries
raw Reference
derived DerivedSeries
resample NumericResampleSeries
resampleV2 NumericResampleSeriesV2
nthPointDownsample NumericNthPointDownsampleSeries
largestTriangleThreeBuckets NumericLargestTriangleThreeBucketsSeries
signalFilter SignalFilterSeries
timeDifference TimeDifferenceSeries
absoluteTimestamp AbsoluteTimestampSeries
timeRangeFilter NumericTimeRangeFilterSeries
timeShift NumericTimeShiftSeries
timeShiftV2 NumericTimeShiftSeriesV2
unitConversion UnitConversionSeries
unitOverride UnitOverrideSeries
valueDifference ValueDifferenceSeries
filterTransformation NumericFilterTransformationSeries
filterWhere NumericFilterWhereSeries
thresholdFilter NumericThresholdFilterSeries
approximateFilter NumericApproximateFilterSeries
dropNan NumericNanFilter
select1dArrayIndex SelectIndexFrom1dNumericArraySeries
aggregateUnderRanges AggregateUnderRangesSeries
rangeDuration RangeDurationSeries
groupWithinRangesV2 GroupWithinRangesSeriesV2
filterByExpression FilterByExpressionSeries
scalarUdf ScalarUdfSeries
enumToNumeric EnumToNumericSeries
contains ContainsSeries
refprop RefpropSeries
refpropV2 RefpropSeriesV2
extractFromStruct ExtractNumericFromStructSeries
tagByIntervals TagByIntervalsSeries
filterByTag NumericTagFilterSeries
selectTags NumericSelectTagsSeries
toStartOfInterval NumericToStartOfIntervalSeries
phaseUnwrap PhaseUnwrapSeries
aggregate AggregateNumericSeries deprecated: Deprecated: AggregateNumericSeries is deprecated. Use numericAggregation with aggregation.GroupByAggregationBuilder instead.
rollingOperation RollingOperationSeries deprecated: Deprecated: RollingOperationSeries is deprecated. Use numericAggregation with aggregation.RollingTimeAggregationBuilder instead.
bitOperation BitOperationSeries deprecated: Deprecated: BitOperationSeries is deprecated in favor of individual nodes for each operation.
channel ChannelSeries deprecated: Deprecated: Deprecated in favor of the SelectSeries node
offset OffsetSeries deprecated: Deprecated: OffsetSeries is deprecated in favor of add with a constant.
scale ScaleSeries deprecated: Deprecated: ScaleSeries is deprecated in favor of multiply with a constant.
arithmetic ArithmeticSeries deprecated: Deprecated: ArithmeticSeries is deprecated in favor of individual nodes for each operation.
unaryArithmetic UnaryArithmeticSeries deprecated: Deprecated: UnaryArithmeticSeries is deprecated in favor of individual nodes for each operation.
binaryArithmetic BinaryArithmeticSeries deprecated: Deprecated: BinaryArithmeticSeries is deprecated in favor of individual nodes for each operation.

NumericStaleRangesV2#

Produces ranges where a numeric input has no data for at least a duration.

input NumericSeries required
threshold compute.api.Duration required: Minimum gap without data that produces an output range.

NumericTagFilterSeries#

Applies a tag predicate to the numeric series produced by input.

input NumericSeries required
predicate TagPredicate required

NumericThresholdFilterSeries#

Outputs only the values of the numeric plot value that satisfy the threshold condition.

input NumericSeries required
threshold DoubleConstant required
operator ThresholdOperator required

NumericTimeRangeFilterSeries#

Filters the series to points within the specified time range.

input NumericSeries required
startTime TimestampConstant: Represents the start time (inclusive) of the time range.
endTime TimestampConstant: Represents the end time (inclusive) of the time range.

NumericTimeShiftSeries#

input NumericSeries required
duration DurationConstant required

NumericTimeShiftSeriesV2#

Shifts the timestamps in the input series by the resolved duration.

input NumericSeries required
duration compute.api.Duration required: Duration to shift the input series by.

NumericToStartOfIntervalSeries#

input NumericSeries required
window compute.api.Duration required

NumericUnionOperation#

DEPRECATED: apply aggregations as a separate step over the unioned output.

MIN: Combines multiple points by taking the minimum value
MAX: Combines multiple points by taking the maximum value
MEAN: Combines multiple points by taking the mean of the values
SUM: Combines multiple points by taking the sum of the values
COUNT: Combines multiple points by taking the count of the values
STANDARD_DEVIATION: Combines multiple points by taking the standard deviation of the values
ROOT_MEAN_SQUARE: Combines multiple points by taking the root mean square of the values
THROW: Throws a DuplicateTimestamp error if two series contain points with a duplicate timestamp.

NumericUnionSeries#

Combines multiple numeric series together and outputs a single series. By default, points from the inputs are concatenated (SQL {@code UNION ALL}) and duplicate timestamps across inputs are preserved. To aggregate values that share a timestamp, do a separate aggregation step next.

input array of NumericSeries: Input series to union
operation NumericUnionOperation deprecated: The strategy to merge points with duplicate timestamps. Deprecated: Apply aggregations as a separate step over the unioned output.

Nyquist#

Returns a complex-valued frequency response, H(f), of the input series to the output series.

input NumericSeries required
output NumericSeries required
stftOptions StftOptions

OffsetSeries#

For every timestamp specified in the input series, offset it by a constant factor.

input NumericSeries required
scalar DoubleConstant required: The constant to add to each point

OnChangeRanges#

Produces a list of ranges for each point that has a different value to the previous point. A range will have identical start and end times.

input Series required

OrSeries#

Combines two boolean series point-wise using logical OR, producing true where either input is true and false otherwise.

left BooleanSeries required
right BooleanSeries required

OutputFormat#

ARROW_V4
ARROW_V3
LEGACY

OutputFrequencyType#

LINEAR: Frequency in Hertz.
ANGULAR: Frequency in radians per second.
NORMALIZED_LINEAR: Normalized frequency, in the range [0, 0.5] for real-valued signals.
NORMALIZED_ANGULAR: Normalized angular frequency, in the range [0, π] for real-valued signals.

OutputPhaseUnit#

RADIANS
DEGREES

OutputRangeStart#

Set type to the name of one of the other fields, and set only that field.

firstPointMatchingCondition FirstPointMatchingCondition: The first point in the output range will be the first point where the condition evaluates to true.
afterPersistenceWindow AfterPersistenceWindow: The output range will contain points where the condition has been true for at least the min duration and min points.

PaddedRanges#

Expands ranges durations by adding temporal padding to the start, end, or both sides. Note that if the input ranges contain a RangeValue, it will be dropped.

input RangeSeries required
padding DurationConstant required
paddingConfiguration RangePaddingConfiguration: Configuration for how to apply padding to the ranges. Defaults to PAD_START_AND_END if not specified.

PaddedRangesV2#

Expands ranges using an authored Duration expression and a selected padding direction.

input RangeSeries required
duration compute.api.Duration required: How far to extend each range on every padded side.
paddingConfiguration RangePaddingConfiguration required

PagedLogPlot#

timestamps array of Timestamp
values array of LogValue
nextPageToken PageToken: Continuation token for this paged result. Can only be returned if paging was requested. To retrieve the next page, repeat the same request with this value as pageToken and keep the pageSize sign unchanged. Returned only when this response contains exactly abs(pageSize) points; the next response may still be empty if this page ended exactly at the result-set boundary. Empty means this response was short and there are no further points in the requested direction.

PageInfo#

Specification of a page for a series. Returns full resolution points beginning nearest to the given page token, advancing pageSize points in the time direction specified by the sign of the page size. Throws if the absolute pageSize is greater than 5,000. Paging cursors are not guaranteed to distinguish exact duplicate non-log points with the same timestamp, tags, and value. If those duplicates straddle a page boundary, later duplicates may be skipped.

pageToken PageToken
pageSize integer required

PageStrategy#

Set type to the name of one of the other fields, and set only that field.

pageInfo PageInfo

PageToken#

Set type to the name of one of the other fields, and set only that field.

timestampAndId TimestampAndId

ParameterInput#

variables map of VariableName to VariableValue
timeRange Range: Overrides the start and end time of the compute request. If either the start or end are not present, we default back to the start/end specified in the request.

ParameterizedContext#

parameterInputs array of ParameterInput: Each parameter input provides a satisfying set of values for the parameterized compute node.

PeakRanges#

Produces a list of ranges for each point that is greater than its neighbors. Peaks at edges are discarded, and continuous, multivalue, flat peaks will return all values.

input NumericSeries required
returnsPeaks boolean deprecated: True if returning peaks, else troughs. Deprecated: No longer used, use returnType instead
returnType PeakType: Optional for backcompatibility.
minimumProminence DoubleConstant: The minimum topographic prominence for an extrema to be returned. Prominence is the minimum vertical distance needed to travel from an extrema to one of greater magnitude.

PeakType#

PEAK
TROUGH
BOTH

PercentageThreshold#

Threshold defined as the percentage of a given value.

value DoubleConstant required: Allowed spread, as a percentage of the window’s mean value.

Percentile#

The value at the specified percentile.

percentile DoubleConstant required: Percentile to compute, in range [0, 1]. E.g. 0.5 for median, 0.95 for p95.

Periodogram#

Returns the spectral density estimate (i.e. PSD) of the input series.

input NumericSeries required
method PeriodogramMethod required

PeriodogramMethod#

WELCH_BLACKMAN
WELCH_HANN
WELCH_HAMMING
WELCH_RECT

PersistenceWindowConfiguration#

Configures how long a condition has to be true for to output a time range, and what to use as the start of the output range. For a point to be included in the output time range, both the minPoints and minDuration conditions must be satisfied.

minPoints IntegerConstant: The minimum number of points for which this condition be must satisfied to include the time range in the output. Must be non-negative. If not present, will default to 1.
minDuration DurationConstant: The minimum number of points for which this condition must be satisfied to include the time range in the output. Must be non-negative. If not present, will default to 1 nanosecond.
outputRangeStart OutputRangeStart required: Which point to use as the start of the output range. Defaults to firstPointMatchingCondition if not specified.

PhaseUnwrapSeries#

Removes discontinuities from a wrapping signal (e.g. phase or heading) by adding whole multiples of period whenever the step between consecutive samples exceeds half the period. Output length equals input length.

input NumericSeries required
period DoubleConstant required: The wrapping period of the signal (e.g. 2*pi for radians, 360 for degrees). A jump between consecutive samples greater than period/2 in magnitude is corrected by adding/subtracting a whole multiple of period.

PhysicalConstant#

Physical constants referenceable by name. The server resolves the symbol to its numeric value in SI units.

SPEED_OF_LIGHT: Speed of light in vacuum, 299792458 m/s (exact).
AVOGADRO: Avogadro constant, 6.02214076e23 mol^-1 (exact).
BOLTZMANN: Boltzmann constant, 1.380649e-23 J/K (exact).
PLANCK: Planck constant, 6.62607015e-34 J*s (exact).
ELEMENTARY_CHARGE: Elementary charge, 1.602176634e-19 C (exact).
GAS_CONSTANT: Molar gas constant, 8.31446261815324 J/(mol*K) (exact, Avogadro times Boltzmann).
STANDARD_GRAVITY: Standard acceleration of gravity, 9.80665 m/s^2 (exact by convention).
STANDARD_ATMOSPHERE: Standard atmosphere, 101325 Pa (exact by convention).
ABSOLUTE_ZERO_CELSIUS: Absolute zero expressed in degrees Celsius, -273.15 (exact).
GRAVITATIONAL_CONSTANT: Newtonian constant of gravitation, 6.67430e-11 m^3/(kg*s^2) (CODATA 2022, measured).
STEFAN_BOLTZMANN: Stefan-Boltzmann constant, 5.670374419e-8 W/(m^2*K^4) (exact derived).

PinnedModuleVersionReference#

version StringConstant required

Point3d#

x number required
y number required
z number required
timestamp Timestamp required

PolynomialCurve#

degree IntegerConstant required: The highest allowable degree of the fit polynomial.
intercept DoubleConstant: The y-value at the point x (or t) = 0. If omitted, the y-intercept will also be fit to the data.

PolynomialCurveFit#

A polynomial fit of the enclosing target (y) against the chosen independent variable (x).

independentVariable CurveFitVariable required: The x axis the target is fit against — either relative sample time or another series.
degree IntegerConstant required: The highest allowable degree of the fit polynomial.
intercept DoubleConstant: The y-value at the point x (or t) = 0. If omitted, the y-intercept will also be fit to the data.

PolynomialResultDetails#

y = sum( a_i * x^i ) for i in 0…len(a)

a array of number

Power#

Raises base to the power of exponent pointwise.

base NumericSeries required: Base values.
exponent NumericSeries required: Exponent values.

PowerCurve#

An empty JSON object, {}.

PowerResultDetails#

y = a * x^b

a number required
b number required

ProductSeries#

For every timestamp specified in the input series, outputs a value that is the product for that timestamp across all input series. Only outputs timestamps where all input series have an entry for that timestamp, subject to interpolation.

inputs array of NumericSeries: Input series across which to take the pointwise product
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

PropertiesFilter#

name string required
values array of string

Property#

name StringConstant required
value StringConstant required

PropertyComparisonOperator#

EQ
NEQ
GT
GTE
LT
LTE

Psd#

input NumericSeries required
stftOptions StftOptions
magnitudeScaling MagnitudeScaling: The scaling to apply to the output magnitude. Defaults to MAGNITUDE_DB_10 if not specified.
outputFrequencyType OutputFrequencyType: The type of the output frequency. Defaults to LINEAR if not specified. Changing the output frequency unit may also rescale the magnitude of the output in order to ensure the density of the output is consistent.
summarizationStrategy FrequencySummarizationStrategy: Strategy to downsample the output frequency spectrum.

QuotaIntervalUsage#

durationSeconds integer (int64) required: Length of the quota window in seconds.
currentWindow QuotaWindow: Bounds of the current quota window. Usage counters reset at the window’s end.
resources array of QuotaResourceUsage: Usage per limited resource. Resources without a configured limit are not reported.

QuotaResourceType#

Resource types that can be limited by the per-user compute quota.

READ_ROWS
READ_BYTES
QUERIES
ERRORS
EXECUTION_TIME

QuotaResourceUsage#

resource QuotaResourceType required
used number required: Amount consumed within the current window, in the resource’s native unit (rows, bytes, query count, or seconds for EXECUTION_TIME).
limit number required: The cap for this resource in this window.

QuotaWindow#

startTime string (date-time) required
endTime string (date-time) required

Range#

The end represents the first timestamp that does not belong to the range. If absent, there is no known end to the range.

start Timestamp
end Timestamp
value RangeValue

RangeAggregation#

An aggregation value representing combined aggregation metrics for data over a range.

average number required
min number required
max number required
standardDeviation number required
count number required
sum number required
rootMeanSquare number

RangeAggregationOperation#

Set type to the name of one of the other fields, and set only that field.

sum Summation
rootMeanSquare RootMeanSquare
average Average
min Minimum
max Maximum
standardDeviation StandardDeviation
count Count
all Empty

RangeDurationSeries#

Outputs a numeric series with one point per range: timestamp = range start, value = range duration.

input RangeSeries required
unboundedRangeBehavior UnboundedRangeBehavior required: How to handle ranges with a missing start or end.
timeUnit TimeUnit: The time unit of the duration. Defaults to seconds if not specified.

RangeIntervalSource#

Intervals come from a range series. Must resolve to literal ranges.

ranges RangeSeries required
tagBy array of RangeTagSource: Non-empty ordered list of tag sources. Each contributes a tag key-value pair per range interval.

RangeMap#

start DoubleConstant: Inclusive start value. If not specified, the start is the prior range’s end value, or negative infinity.
end DoubleConstant: Exclusive end value. If not specified, the end value is the next range’s start value, or positive infinity.
output StringConstant required: The value to map to if the input value is within the range.

RangePaddingConfiguration#

PAD_START
PAD_END
PAD_START_AND_END

RangesAnchor#

Uses the start timestamp of the range at the configured zero-based index from ranges evaluated over each interval, considering only ranges whose start lies in the anchor window [start, end), or [start, +infinity) when the end is absent. The window is the run bounds for a run anchor. A range already in progress when the window opens anchors at the window start. Series in ranges that select no explicit dataset resolve against the aligned input dataset, and the computed range groups must carry the aligned runs’ runRid grouping tags.

ranges RangeSeries required: The range series evaluated per interval to locate the anchor range.
index IntegerConstant: Zero-based range index after sorting. Defaults to zero and must be non-negative.
sortOrder RangeSortOrder: Ordering by range start timestamp before selecting the index. Defaults to ASC.

RangeSeries#

Set type to the name of one of the other fields, and set only that field.

approximateThreshold ApproximateThresholdRanges
booleanRanges BooleanRanges
thresholdRangesV2 ThresholdRangesV2
durationFilter DurationFilterRanges
durationFilterV2 DurationFilterRangesV2
enumFilter EnumFilterRanges
categoryFilterRangesV2 CategoryFilterRangesV2
categoricalOnChangeRangesV2 CategoricalOnChangeRangesV2
categoricalStaleRangesV2 CategoricalStaleRangesV2
categoricalEqualityRangesV2 CategoricalEqualityRangesV2
enumSeriesEqualityRangesNode EnumSeriesEqualityRanges
eventsSearch EventsSearchRanges
intersectRange IntersectRanges
literalRanges LiteralRanges
minMaxThreshold MinMaxThresholdRanges: Computes ranges where the input time series matches a filter defined by lower and upper bounds, and an operator.
not NotRanges
onChange OnChangeRanges
peak PeakRanges
extremaRangesV2 ExtremaRangesV2
rangeNumericAggregation RangesNumericAggregation
raw Reference
derived DerivedSeries
seriesCrossoverRangesNode SeriesCrossoverRanges
seriesEqualityRangesNode SeriesEqualityRanges
numericEqualityRangesV2 NumericEqualityRangesV2
stabilityDetection StabilityDetectionRanges
stabilityDetectionV2 StabilityDetectionRangesV2
staleRange StaleRanges
numericStaleRangesV2 NumericStaleRangesV2
threshold ThresholdingRanges: Computes ranges where the input time series matches a filter defined by a single threshold and an operator.
unionRange UnionRanges
paddedRanges PaddedRanges
paddedRangesV2 PaddedRangesV2

RangesNumericAggregation#

Aggregates the values of a numeric series at each range specified by the input ranges.

ranges RangeSeries required
input NumericSeries required
operation RangeAggregationOperation required

RangeSortOrder#

Sort order for selecting ranges. ASC selects from earliest to latest, DESC selects from latest to earliest.

ASC
DESC

RangesSummary#

A summary of the ranges returned from a SummarizeRanges request. Returned when the number of ranges found is above a threshold.

rangesSummary array of RangeSummary

RangeStart#

Selects which point of a satisfied condition begins the output range.

CONDITION_START
PERSISTENCE_START

RangeSummary#

Summary of a set of ranges

range Range required: The range representing the minimum start and maximum end times of the ranges.
subRangeCount integer required: The number of ranges found within this time range.

RangeTagSource#

A dimension contributing a tag key-value pair per range interval.

Set type to the name of one of the other fields, and set only that field.

intervalIndex IntervalIndexTag
rangeStartTimestamp IntervalStartTimestampTag

RangeValue#

A value attached to a range. If there are no relevant data points in the range, this will be noPointsInRange for the range; otherwise, an Aggregation will be returned if this is produced by getting multiple aggregation values over a range (i.e. aggregating for all metrics over a range), and will be a double otherwise for singular values (i.e. aggregating for just the sum of data values over a range).

Set type to the name of one of the other fields, and set only that field.

double number
aggregation RangeAggregation
noPointsInRange Empty

Reference#

name string required: A reference to a value not defined in the current compute scope.

RefpropProperty#

Supported properties for REFPROP calculations.

TEMPERATURE
PRESSURE
MASS_DENSITY
MASS_SPECIFIC_ENTHALPY
MASS_SPECIFIC_INTERNAL_ENERGY
MASS_SPECIFIC_ENTROPY
VISCOSITY

RefpropSeries#

Computes the output property given input properties for the substance using REFPROP. Initial implementation only expects two input properties.

inputs map of string to NumericSeries: A map that maps a REFPROP property to its numeric series.
outputProperty RefpropProperty required: The desired output property. This should not be one of the input properties.
substance RefpropSubstance required: The substance for REFPROP calculations.
interpolationConfiguration InterpolationConfiguration deprecated: Deprecated: Setting interpolation on this node is deprecated.

RefpropSeriesV2#

Computes a REFPROP output property from the supplied substance and output property.

inputs map of string to NumericSeries: Map of REFPROP property to its numeric series.
substance RefpropSubstance required: The substance for REFPROP calculations.
outputProperty RefpropProperty required: The desired output property. This must not be one of the input properties.

RefpropSubstance#

Supported substances for REFPROP calculations

NITROGEN
OXYGEN
HELIUM
HYDROGEN
METHANE
PROPANE
BUTANE
WATER
XENON
AIR
CARBON_DIOXIDE
ARGON
CARBON_MONOXIDE
JET_A_4658
JET_A_3638

RequestInactive#

No query for this request was running at sample time; not necessarily finished.

An empty JSON object, {}.

RequestInProgress#

Summed across the request’s active queries when it fans out into more than one.

readRows integer (int64) required
totalRowsApprox integer (int64) required: May grow while a query runs as new sources to process become known.

RequestProgress#

Set type to the name of one of the other fields, and set only that field.

inProgress RequestInProgress
inactive RequestInactive

ResourceSearchQuery#

Predicate tree for filtering.

Set type to the name of one of the other fields, and set only that field.

properties PropertiesFilter
numericProperty NumericPropertyPredicate
numericPropertyRange NumericPropertyRangePredicate: Matches resources using an inclusive numeric property range.
labels LabelsFilter
workspace StringConstant
and array of ResourceSearchQuery
or array of ResourceSearchQuery
not ResourceSearchQuery

RollingOperationSeries#

DEPRECATED. Use numericAggregation with aggregation.RollingTimeAggregationBuilder instead.

input NumericSeries required
window Window required
operator NumericAggregationOperator required

RollingPointsAggregationBuilder#

Rolling point-count aggregation: for each point in the input, aggregate the current point and the preceding N - 1 points within the same tag grouping.

input Series required
points IntegerConstant required

RollingTimeAggregationBuilder#

Rolling time-window aggregation: for each point in the input, aggregate the values within the preceding time window.

input Series required
window compute.api.Duration required

RootMeanSquare#

The root mean square of points inside the time window.

An empty JSON object, {}.

Run#

Run reference for Dataset.run. The rid field accepts either a literal run RID or a StringConstant reference resolved from Context.

rid StringConstant required: Resource identifier of a run

run.api.Duration#

seconds integer (int64) required
nanos integer (int64) required

RunChannel#

runRid StringConstant required
assetRid StringConstant: Used to disambiguate when multiple assets within this run contain data scopes with the same name. If not specified for a run with multiple assets, an error will be thrown.
dataScopeName StringConstant required: Used to disambiguate when multiple data scopes within this run contain channels with the same name.
channel StringConstant required
additionalTags map of string to StringConstant deprecated: Deprecated: Use additionalTagFilters
additionalTagFilters TagFilters: Tags to filter the channel by, in addition to tag filters defined for a given Run data scope. Throws on collisions with tag keys already defined for the given Run data scope. Only returns points that match both sets of tag filters. For log series, include arg filters here in addition to tag filters.
tagsToGroupBy array of string deprecated: Deprecated: Use groupByTags instead. Throws if both tagsToGroupBy and groupByTags are specified.
groupByTags array of StringConstant: Tags that the channel should be grouped by. If this is non-empty a grouped result will be returned with an entry for each grouping. Only one of tagsToGroupBy and groupByTags should be specified.

RunsSearchTarget#

Search runs. Each matched run expands to one grouping per data source attached to the run: asset-typed data sources contribute one grouping per data scope on the asset; dataset-typed data sources contribute one grouping per dataset. Every grouping is tagged by runRid and dataScope (the asset’s data scope name for asset-typed sources, or the run-local data source ref name for dataset-typed sources); asset-typed groupings additionally carry assetRid.

query ResourceSearchQuery required: Predicate tree used to filter matching runs.

RustUdfSource#

Rust source code defining the UDF entrypoint and any helper functions.

sourceCode string required: Rust source code to transpile into WASM. The operation consuming this source defines the entrypoint signature: scalar UDFs receive one scalar per input series, while aggregation UDFs receive grouped value vectors. If several functions appear at the top level, annotate the entrypoint with #[main].

Sampling#

Options for returning a sampled subset of raw series points.

Set type to the name of one of the other fields, and set only that field.

everyNthPerSeries EveryNthPerSeriesSampling
hashSubsample HashSubsampleSampling

SavedDataset#

A persisted source of data, named by its RID. Can be a catalog dataset, pipeline, connection, etc.

rid StringConstant required: A data source RID, given as a literal or as a variable resolved from Context. A pipeline’s logic only accepts catalog datasets/pipelines, since it is persisted without context.

ScalarUdfSeries#

Executes user-provided scalar UDF code over aligned inputs. The function must return one output for each aligned input row, with the return type matching the series type that contains this node.

inputs array of Series: Input series passed to the UDF in argument order.
source UdfSource required: Language-specific source definition for the UDF.

ScaleSeries#

For every timestamp specified in the input series, multiply it by a constant factor.

input NumericSeries required
scalar DoubleConstant required: The constant to multiply each point by
scalarUnit string: The units of the scalar to multiply by. If empty, the scalar is considered unit-less.

Scatter#

A scatter plot comprised of the values from one input series on the x-axis and values from another input series on the y-axis, for all points within a requested time range.

x NumericSeries required
y NumericSeries required

Scatter3d#

x NumericSeries required
y NumericSeries required
z NumericSeries required
driverSeries DriverSeries3d

ScatterCurveFit#

Fits a curve to the scatter between two series.

scatter Scatter required: The x and y series to fit to.
options ScatterFitOptions required

ScatterFitOptions#

minX DoubleConstant: Inclusive lower bound on the x values to fit to. If omitted, does not set a bound (equivalent to setting this to the minimum x value among all points in the time range).
maxX DoubleConstant: Inclusive upper bound on the x values to fit to. If omitted, does not set a bound (equivalent to setting this to the maximum x value among all points in the time range).
minY DoubleConstant: Inclusive lower bound on the y values to fit to. If omitted, does not set a bound (equivalent to setting this to the minimum y value among all points in the time range).
maxY DoubleConstant: Inclusive upper bound on the y values to fit to. If omitted, does not set a bound (equivalent to setting this to the maximum y value among all points in the time range).

ScatterSummarizationStrategy#

Set type to the name of one of the other fields, and set only that field.

spatial SpatialDecimateStrategy
temporal TemporalDecimateStrategy

ScatterTemporalAggregation#

MEAN
MIN
MAX
FIRST
LAST

SearchDataset#

Resolves matching assets or runs from search, creating a combined dataset tagged by source rid and data scope.

target SearchTarget required: Whether to search assets or runs and with which query.
maxResults integer: Maximum number of assets or runs returned from search. Defaults to 10000 when omitted.

SearchEventOriginType#

WORKBOOK
TEMPLATE
API
DATA_REVIEW
PROCEDURE

SearchTarget#

Set type to the name of one of the other fields, and set only that field.

assets AssetsSearchTarget
runs RunsSearchTarget

SelectIndexFrom1dEnumArraySeries#

For each timestamp, selects a single enum value from the 1D enum array at the specified index. If the index is out of bounds for an array at a given timestamp, it is omitted.

input Enum1dArraySeries required
index IntegerConstant required

SelectIndexFrom1dNumericArraySeries#

For each timestamp, selects a single numeric value from the 1D numeric array at the specified index. If the index is out of bounds for an array at a given timestamp, it is omitted.

input Numeric1dArraySeries required
index IntegerConstant required

SelectNewestPointsSeries#

Select the most recent N points from the input series by timestamp.

input NumericSeries required: The input series to select the newest points from.
numPoints IntegerConstant required: The number of points to select.

SelectOldestPointsSeries#

Select the earliest N points from the input series by timestamp.

input NumericSeries required: The input series to select the oldest points from.
numPoints IntegerConstant required: The number of points to select.

SelectProperty#

Selects a typed property value from every asset or run branch resolved from dataset. Inside a WithSeriesDataset definition, dataset may be omitted to use that node’s wrapping input; it is required everywhere else. Asset branches read properties directly from their asset. Run branches read properties directly from their run, never from a backing asset or dataset. Direct saved-dataset branches are unsupported.

propertyName string required
dataset Dataset

SelectSeries#

Selects series from a dataset (might contain multiple tag groupings). When name is provided, selects that single channel. When name is omitted, selection starts from all channels in the dataset scope; callers narrow the selection with a filterByTag predicate on the reserved NOMINAL_CHANNEL tag key and preserve per-channel identity in the response with selectTags. Providing both name and a NOMINAL_CHANNEL filter is rejected as an invalid request. Raw and derived channels can both be selected through the reserved tag filter.

name StringConstant: The identifying series name to query
dataset Dataset: The dataset providing the data to query. May be omitted inside a WithSeriesDataset series definition, in which case it defaults to that node’s wrapping input; required everywhere else.
sampling Sampling: Raw-series sampling options. Supported for numeric and enum raw series.

SelectValue#

Set type to the name of one of the other fields, and set only that field.

firstPoint Series
firstValuePoint Series
firstRange RangeSeries
lastPoint Series
lastValuePoint Series
lastRange RangeSeries
nthRange NthRange

SerializableError#

A SerializableError is a representation of a ServiceException that exists to send error results to clients as part of a response object when directly throwing an exception is undesirable.

name string required
message string
errorInstanceId string required
statusCode integer required
params map of string to string

Series#

Set type to the name of one of the other fields, and set only that field.

raw Reference deprecated: Deprecated: use typed value of series instead.
enum EnumSeries
numeric NumericSeries
log LogSeries
array ArraySeries
struct StructSeries
video Reference

SeriesCrossoverRanges#

Produces a list of zero-duration ranges at the first point where two series cross over one another

inputA NumericSeries required
inputB NumericSeries required: The series inputA is compared against.

SeriesEqualityRanges#

Produces a list of ranges for which provided series are all equal (or are not all equal).

input array of NumericSeries
equalityOperator EqualityOperator required
tolerance DoubleConstant
persistenceWindowConfiguration PersistenceWindowConfiguration
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

SeriesVariable#

A series selected from a dataset, bound to a variable. The series type (numeric, enum, …) is decided lazily by the node consuming the variable, matching the behaviour of channel variables.

select SelectSeries required
groupByTags array of StringConstant: Tags that the series should be grouped by. If this is non-empty a grouped result will be returned with an entry for each grouping; otherwise all matching series are merged.

SetNegativeValuesToZero#

An empty JSON object, {}.

SetOperator#

AND
OR

SignalFilterConfiguration#

Set type to the name of one of the other fields, and set only that field.

lowPass LowPassConfiguration
highPass HighPassConfiguration
bandPass BandPassConfiguration
bandStop BandStopConfiguration

SignalFilterSeries#

Applies IIR-based signal filtering to input series. Includes low-pass, high-pass, band-pass, and band-stop filters. Currently supports variable-order bidirectional Butterworth filters, with fixed-size padding based on SciPy output.

input NumericSeries required
signalFilterConfiguration SignalFilterConfiguration required: Filter type and cutoff frequencies.
order IntegerConstant required: Order of filter. Must be a positive integer, and is effectively doubled for bidirectional filters.
samplingFrequency DoubleConstant: The sampling frequency of the input series. Used to calculate normalized frequency for cutoff frequencies.

Sin#

Computes the trigonometric sine of each value (input in radians).

x NumericSeries required: Input numeric series (radians).

SinglePoint#

Return type representing a single point value.

timestamp Timestamp required
value Value required
precisionLoss boolean required: Returns true if the result required downcasting to a type with less precision, for example if the input series was Int64 and the result is a Float64.

SpatialDecimateStrategy#

Decimate by spatial trees. Creates buckets by bisecting on each dimension, creating quadrants for 2d scatter and octants for 3d scatter. Continues subdividing by prioritizing larger undivided buckets until reaching the max amount of buckets.

An empty JSON object, {}.

Spectrogram#

The PSD or amplitude of the input over time, from a short-time Fourier transform of overlapping windows.

input NumericSeries required: A numeric series from a single data source. Series grouped by tags are not supported.
stftOptions StftOptions: Window and FFT parameters.
quantity SpectrogramQuantity: What each cell measures. Defaults to PSD.
magnitudeScaling MagnitudeScaling: How each cell’s value is scaled. Defaults to decibels: 10·log10 of the PSD, or 20·log10 of the amplitude.
bucketing SpectrogramBucketing: The output’s time and frequency buckets, frequency band and reductions.

SpectrogramBucketing#

The output’s time and frequency buckets: at most maxTimeBuckets rows and maxFrequencyBuckets columns over the frequency band, with the values that share a row or column combined by timeReduction and frequencyReduction. Invalid values fail with InvalidExpression.

maxTimeBuckets integer: Maximum number of rows, further lowered so that rows × columns is at most 262,144. Defaults to 256.
maxFrequencyBuckets integer: Maximum number of columns. Defaults to 512.
minFrequency number: Inclusive lower bound of the band in Hz. Defaults to 0.
maxFrequency number: Exclusive upper bound of the band in Hz. Defaults to including the Nyquist frequency.
timeReduction SpectrogramReduction: Combines the windows that share a row. Defaults to MEAN.
frequencyReduction SpectrogramReduction: Combines the frequency bins that share a column. Defaults to MEAN.

SpectrogramPlot#

Values on a time-frequency grid. Columns are frequency buckets and rows are the time buckets that hold data, in ascending time.

frequencyCenters array of number: Center frequency in Hz of each column.
frequencyEdges array of number: Column boundaries in Hz, one more than frequencyCenters: column c spans frequencyEdges[c] to frequencyEdges[c + 1]. Columns may differ in width.
rows array of SpectrogramRow

SpectrogramQuantity#

What a spectrogram cell measures.

PSD: Power spectral density, in the input unit squared per Hz.
AMPLITUDE: Peak amplitude, in the input unit.

SpectrogramReduction#

How the values that share one output row or column are combined, before scaling.

MEAN: Arithmetic mean.
MAX: Maximum, which keeps short transients visible.
GEOMETRIC_MEAN: Geometric mean, which suits decibel displays with a wide dynamic range.

SpectrogramRow#

start Timestamp required: Start of the time span this row covers.
end Timestamp required: End of the time span this row covers.
values array of number: One value per column

Sqrt#

Computes the square root of each value.

x NumericSeries required: Input numeric series (non-negative values).

StabilityDetectionRanges#

Outputs a set of ranges where the input series is stable. For each point, the min and max are calculated over the specified lookback window, including the current point. A point is considered stable if its value does not deviate from the calculated min and the max by more than the threshold and the total number of points within the window is at least the specified amount. The threshold can be either fixed values or percentages of the value. The lookback window must be strictly positive. The minimum points threshold defaults to 2.

input NumericSeries required
windowConfiguration StabilityWindowConfiguration required
threshold Threshold required

StabilityDetectionRangesV2#

Produces ranges where a numeric input is stable within a modern duration window.

input NumericSeries required
window compute.api.Duration required: Lookback window over which the input’s spread is measured.
minPoints IntegerConstant: Minimum number of points in the window. Defaults to 2.
threshold Threshold required

StabilityWindowConfiguration#

minPoints IntegerConstant: The minimum number of points within the window to create a stable range. Must be non-negative. If not present, will default to 2.
duration DurationConstant required

StaleRanges#

Produces a list of ranges for which data does not exist for the specified duration or longer. Increases window size by the specified staleness threshold on both ends to capture edge cases of data not currently in view.

input Series required
threshold DurationConstant required
startTimestamp TimestampConstant: The start timestamp of the range. If not specified, staleness will automatically use view range start.
endTimestamp TimestampConstant deprecated: Deprecated: No longer used and will be removed

StandardDeviation#

The standard deviation of points inside the time window.

An empty JSON object, {}.

StftOptions#

Options for frequency analyses that are backed by short-time Fourier transforms.

window FftWindow: Window function type to use
windowSymmetry WindowSymmetry: Either DFT_EVEN or SYMMETRIC
samplingFrequency number: Sampling frequency of the series in Hz. If absent, will calculate using the input series.
nperseg integer: Number of points per window segment.
noverlap integer: Number of points of overlap between windows. Must be less than nperseg.
nfft integer: Number of points to use in the FFT. Must be greater than or equal to nperseg.
detrend DetrendType: Detrend the points in each window segment. If absent, will use the default detrend for that function.

StringConstant#

A string constant that can be a literal value or a named variable reference.

Set type to the name of one of the other fields, and set only that field.

literal string: Create from a literal string value.
variable string: Create from a named reference.

StringPropertyDimension#

Injects the value of a string property, under a tag key equal to the property name by default. Only supported on asset and run branches: asset branches read the asset’s properties, and run branches, including run branches backed by an asset, read the run’s properties, matching SelectProperty. Values are injected as is. A branch without the property, or whose property holds a numeric value, gets the reserved value NOMINAL_ABSENT.

propertyName StringConstant required: Name of the property whose value to inject. Also the default tag key.

StringSetConstant#

Set type to the name of one of the other fields, and set only that field.

literal array of string
variable string

StringSetConstantV2#

A set of string constants that can be a literal set or a named variable reference.

Set type to the name of one of the other fields, and set only that field.

literal array of StringConstant: Create from a literal set of string constants.
variable string: Create from a named reference.

StructSelectTagsSeries#

Subsets the tag dimensions of the input struct series to the given tag keys. Preserves the input series’ cardinality: points that share the selected keys but differ on dropped keys remain distinct.

input StructSeries required
tagKeys array of StringConstant

StructSeries#

Set type to the name of one of the other fields, and set only that field.

channel ChannelSeries
raw Reference
derived DerivedSeries
combine CombineStructSeries
selectStruct SelectSeries
extractFromStruct ExtractStructFromStructSeries
toStartOfInterval StructToStartOfIntervalSeries
timeShift StructTimeShiftSeries
filterByTag StructTagFilterSeries
selectTags StructSelectTagsSeries

StructTagFilterSeries#

Applies a tag predicate to the struct series produced by input.

input StructSeries required
predicate TagPredicate required

StructTimeShiftSeries#

input StructSeries required
duration DurationConstant required

StructToStartOfIntervalSeries#

input StructSeries required
window compute.api.Duration required

Subtract#

Subtracts right from left pointwise.

left NumericSeries required: Left-hand operand (minuend).
right NumericSeries required: Right-hand operand (subtrahend).

SummarizationStrategy#

Set type to the name of one of the other fields, and set only that field.

decimate DecimateStrategy
largestTriangleThreeBuckets LttbStrategy: Largest-Triangle-Three-Buckets downsampling returns selected numeric timestamp/value points that preserve the visual shape of each group. It is distinct from fixed-width bucket aggregation.
page PageStrategy: Paging returns full resolution points for supported series types. Results include a continuation token when another request may be needed to continue in the requested direction.
truncate TruncateStrategy: Truncation returns a bounded set of full-resolution points. For compute requests, it is supported for array and struct series. To retrieve full-resolution numeric series, use paging.

SummarizeCartesian#

input Cartesian required
bounds CartesianBounds
summarizationStrategy ScatterSummarizationStrategy: The summarization strategy to use when there are more than maxPoints. Defaults to spatial.
maxPoints integer: The maximum number of points to return in the response. If more points are found, a BucketedCartesianPlot will be returned. Maximum is 10,000. Defaults to 2,000 if not specified.

SummarizeCartesian3d#

input Cartesian3d required
bounds Cartesian3dBounds
maxPoints integer: The maximum number of points to return in the response. If more points are found, a BucketedCartesian3dPlot will be returned. Maximum is 10,000. Defaults to 2,000 if not specified.
summarizationStrategy ScatterSummarizationStrategy: The strategy to use when summarizing the series. Only spatial decimation is supported.

SummarizeMultivariate#

inputs array of MultivariateInput: List of input series to temporally align at the raw point level, then bucket the aligned tuples together. The series with the largest tag group acts as the “driver” that determines which raw timestamps are included in the alignment. All subsequent series are aligned to the driver’s raw timestamps using the interpolation configuration. After alignment, the resulting (x, y, z, …) tuples are bucketed into time windows and aggregated.
outputFormat OutputFormat: The output format of the response. Defaults to LEGACY (JSON).
bucketCount integer: The number of buckets to return in the response. Maximum is 10,000. Defaults to 1,000 if not specified.
interpolationConfiguration InterpolationConfiguration: Controls how the N input series are aligned at the raw point level when producing multivariate buckets. When present, the server may fill forward a series value from the most recent prior raw point (bounded by the interpolation radius). This alignment happens BEFORE bucketing, ensuring that aggregated values in each bucket are computed from temporally aligned raw tuples. When absent, the default interpolation configuration is used.

SummarizeRanges#

input RangeSeries required
maxRanges integer: The maximum number of ranges to return in the response. If more ranges are found, a RangesSummary will be returned. Defaults to 2000 if not specified.

SummarizeSeries#

Summarizes the output of a series node. The output can be a numeric, enum, log, or cartesian series. Summarization strategy should be specified.

input Series required
outputFormat OutputFormat: The output format of the response. Defaults to LEGACY.
numericOutputFields array of NumericOutputField deprecated: The fields to output from the summarization. Applies only to Arrow format numeric series. Deprecated: Use numericOutputFieldsV2 instead.
numericAggregations map of string to NumericAggregationOperator: Additional numeric aggregations per decimation bucket (e.g. percentiles). Map key is the name of the column in the result.
numericOutputFieldsV2 array of NumericOutputFieldV2: Numeric bucket output fields to return. Applies only to Arrow format numeric series. Defaults to MEAN when no numeric output selector is specified. Mutually exclusive with numericOutputFields and numericAggregations.
resolution integer (int64) deprecated: Resolution of the output series specifying time interval between decimated points, in nanoseconds. Deprecated: Use summarizationStrategy instead.
buckets integer deprecated: Number of points to generate in the output series. Deprecated: Use summarizationStrategy instead.
summarizationStrategy SummarizationStrategy: The strategy to use when summarizing the series.

Summation#

The sum of points inside the time window.

An empty JSON object, {}.

SumSeries#

For every timestamp specified in the input series, outputs a value that is the sum for that timestamp across all input series. Only outputs timestamps where all input series have an entry for that timestamp, subject to interpolation.

inputs array of NumericSeries: Input series across which to take the pointwise sum
interpolationConfiguration InterpolationConfiguration deprecated: Defaults to forward fill interpolation with a 1s interpolation radius Deprecated: Setting interpolation on this node is deprecated.

TagAnd#

Conjunction of two tag predicates (both must hold).

left TagPredicate required
right TagPredicate required

TagByIntervalsSeries#

Injects tags derived from interval boundaries onto a numeric series. Points within an interval get tagged; points outside all intervals are dropped. Intervals come from either events or ranges. If intervals with different tag map overlap, a data point falling in multiple intervals is duplicated into each matching output group. Overlapping intervals with equal tag maps are not allowed. Optional time shift subtracts a per-interval anchor from timestamps after matching.

input NumericSeries required
intervals IntervalSource required
timeShift TagByIntervalsTimeShift: When set, shift each matched point’s timestamp by the chosen anchor for that interval.

TagByIntervalsTimeShift#

How to shift timestamps after tagging (subtract anchor from each point in the interval slice).

Set type to the name of one of the other fields, and set only that field.

intervalStart IntervalStartTimeShift
intervalEnd IntervalEndTimeShift
eventTimeShift EventTimeShift

TagFilter#

A filter on a single tag key and set of values. For an IN operator, the tag key must match at least one of the values in the set. For a NOT_IN operator, it must not match any of the values in the set.

key StringConstant required
values array of StringConstant
operator TagFilterOperator required

TagFilterOperator#

IN
NOT_IN

TagFilters#

A set of tag filters to apply to a channel. Allows for combining filters on multiple tag keys, but throws if any one tag key is specified in more than one filter.

Set type to the name of one of the other fields, and set only that field.

single TagFilter
and array of TagFilters

TaggedDataset#

Injects one ephemeral tag (key and value) onto the underlying dataset.

input Dataset required: The underlying dataset to tag.
key StringConstant required: Tag key to inject onto each series in the dataset.
value StringConstant required: Tag value to assign to the injected key.

TagIn#

Constrains a tag key to be in the given set of values.

key StringConstant required
values StringSetConstantV2 required

TagNotIn#

Constrains a tag key to not be in the given set of values.

key StringConstant required
values StringSetConstantV2 required

TagPredicate#

Conjunctive tag predicate for dataset filtering. Compiles to resolved TagFilters for storage and resolution-time folding against known tag assignments (e.g. TaggedDataset, search branches).

Set type to the name of one of the other fields, and set only that field.

in TagIn
notIn TagNotIn
and TagAnd

Tan#

Computes the trigonometric tangent of each value (input in radians).

x NumericSeries required: Input numeric series (radians).

TemporalDecimateStrategy#

Buckets the input by time range into equally sized buckets. The number of buckets is determined by the maxPoints parameter. Returns a CartesianPlot.

xAggregation ScatterTemporalAggregation required: The strategy to use for aggregating the x values in each bucket.
yAggregation ScatterTemporalAggregation required: The strategy to use for aggregating the y values in each bucket.

Threshold#

Set type to the name of one of the other fields, and set only that field.

absolute AbsoluteThreshold
percentage PercentageThreshold

ThresholdingRanges#

Produces a list of ranges for which the threshold condition is satisfied.

input NumericSeries required
threshold DoubleConstant required
operator ThresholdOperator required
persistenceWindowConfiguration PersistenceWindowConfiguration

ThresholdOperator#

GREATER_THAN
GREATER_THAN_OR_EQUAL_TO
LESS_THAN
LESS_THAN_OR_EQUAL_TO
EQUAL_TO
NOT_EQUAL_TO

ThresholdRangesComparison#

Selects the comparison applied to a numeric threshold.

Set type to the name of one of the other fields, and set only that field.

gt ThresholdRangesOrderedComparison
ge ThresholdRangesOrderedComparison
lt ThresholdRangesOrderedComparison
le ThresholdRangesOrderedComparison
eq ThresholdRangesEqualityComparison
ne ThresholdRangesEqualityComparison

ThresholdRangesEqualityComparison#

An equality comparison against a numeric threshold, within an optional tolerance.

threshold DoubleConstant required: The value each input point is compared against.
tolerance DoubleConstant: Maximum absolute difference from the threshold still counted as equal. Defaults to zero, i.e. exact equality.

ThresholdRangesOrderedComparison#

An ordering comparison against a single numeric threshold.

threshold DoubleConstant required: The value each input point is compared against.

ThresholdRangesV2#

Produces ranges where a numeric input satisfies a threshold comparison.

input NumericSeries required
minPoints IntegerConstant: Minimum number of matching points. Defaults to 1.
minDuration compute.api.Duration: Minimum matching duration. Defaults to one nanosecond.
rangeStart RangeStart: Selects whether the output begins at the first match or after persistence is satisfied.
comparison ThresholdRangesComparison required

TimeDifferenceSeries#

Outputs a new series where each value is the difference between the time of the current and previous points.

input Series required
timeUnit TimeUnit: The time unit used to define the output values. Defaults to seconds if not specified.

TimeSeriesCurveFit#

Fits a curve to a series vs its timestamps.

series NumericSeries required: The series to fit. Timestamps will be used as x values and data as y values. The leftmost (earliest) timestamp will be used as the value of 0, and all other timestamps will be relative to that.
options TimeSeriesFitOptions required

TimeSeriesFitOptions#

timeUnit TimeUnit: Time unit used for the timestamps to fit against. Defaults to seconds if not specified.

TimeShiftedDataset#

Transforms a dataset by adding a time offset to all its data.

input Dataset required: The underlying dataset to time-shift.
offset compute.api.Duration required: The time shift to apply. Positive values shift data forward in time, negative values shift backward.

Timestamp#

Deprecated for new APIs. Use google.protobuf.Timestamp for absolute instants in proto APIs.

This legacy nanosecond precision timestamp type is represented by an epoch time in seconds and a nanosecond offset. The nanosecond offset is from the start of the epoch second, so must be less than 1 billion.

seconds integer (int64) required
nanos integer (int64) required

TimestampAndId#

timestamp Timestamp required
id string (UUID)

TimestampConstant#

A timestamp constant that can be a literal value or a named variable reference.

Set type to the name of one of the other fields, and set only that field.

literal Timestamp: Create from a literal timestamp value.
variable string: Create from a named reference.

TimeUnit#

DAYS
HOURS
MINUTES
SECONDS
MILLISECONDS
MICROSECONDS
NANOSECONDS

TruncateStrategy#

Set type to the name of one of the other fields, and set only that field.

maxPointsToReturn integer: Maximum number of points to return before truncating, ordered by timestamp. Throws if greater than 10,000

UdfSource#

Source provider for a user-defined function.

Set type to the name of one of the other fields, and set only that field.

rust RustUdfSource

UnaryArithmeticOperation#

COS: Calculates the trigonometric cosine (input in radians)
SIN: Calculates the trigonometric sine (input in radians)
TAN: Calculates the trigonometric tangent (input in radians)
ABS: Calculates the absolute value
ASIN: Calculates the arcsine, returning radians
ACOS: Calculates the arccosine, returning radians
LOG: Calculates the base 10 logarithm
LN: Calculates the natural logarithm (base e)
SQRT: Calculates the square root

UnaryArithmeticSeries#

Applies a point-wise transformation to a series. The transformation function is applied to every individual data point.

input NumericSeries required
operation UnaryArithmeticOperation required

UnboundedBehavior#

Dictates how empty range bounds should be handled by the compute node. Defaults to INFINITY.

INFINITY
WINDOW_BOUND

UnboundedRangeBehavior#

How to handle ranges with a missing start or end.

Set type to the name of one of the other fields, and set only that field.

clampToWindow Empty: Clamp a missing range start to the compute request window start and a missing range end to the compute request window end.
drop Empty: Drop the range if it has a missing start or end.

Union#

Uses the union of all input timestamps as the alignment grid.

An empty JSON object, {}.

UnionRanges#

inputs array of RangeSeries: The ranges to union.

UnitComputationError#

Set type to the name of one of the other fields, and set only that field.

incompatibleUnitsOperation IncompatibleUnitOperation
unitsMissing UnitsMissing
error SerializableError

UnitConversionSeries#

Convert the given series to a different unit.

input NumericSeries required
outputUnit string required

UnitOperation#

UNIT_AS_EXPONENT
NON_INTEGER_AS_EXPONENT
FLOOR_DIVISION
MODULO
CROSSOVER_COMPARISON
MAX
MIN
MEAN
UNION
ADDITION
SUBTRACTION
UNIT_CONVERSION
COMPARISON
TERNARY_OPERATOR

UnitOverrideSeries#

Override the unit of the given series without changing its values. Unlike unit conversion, no compatibility with the input’s current unit is required.

input NumericSeries required
unit string: The unit to report for the input series. If absent, the input’s unit is cleared.

UnitResult#

Set type to the name of one of the other fields, and set only that field.

success string
noUnitAvailable array of UnitComputationError: The resulting values have no units associated. If this is because of an error, it is listed here.

UnitsMissing#

At least one input is missing a unit.

seriesRids array of string (RID) deprecated: Deprecated: Use channels instead.
channels array of DataSourceAndChannel

Value#

A single value. The type is dictated by the returned variant

Set type to the name of one of the other fields, and set only that field.

stringValue string
float64Value number
int64Value string
arrayValue array of Value
structValue any JSON value: A JSON representation of the struct

ValueDifferenceSeries#

Calculates the first discrete difference of the input series. For each point, computes the difference between the current value and the previous value (current - previous). The first point in the input series is omitted from the output since it has no previous point to compare against.

This is equivalent to pandas DataFrame.diff() with period=1. Useful for analyzing changes between consecutive data points in time series data, such as detecting value increases or decreases.

Example: For input values [1.0, 1.0, 2.0, 2.5, 1.8], the output would be [0.0, 1.0, 0.5, -0.7] at the corresponding timestamps (the first timestamp is omitted).

input NumericSeries required
negativeValuesConfiguration NegativeValueConfiguration: Defines the strategy for handling negative output values. Defaults to allowNegativeValues if not specified.

ValueMapSeries#

Maps a continuous numeric series to a discrete enum series using the specified value ranges.

input NumericSeries required: The input series to map to an enumerated series
mapping array of RangeMap: The output of the first capturing range will be used. Ranges are start inclusive, end exclusive, must not overlap, and increasing from lowest to highest. Ranges can be open ended to the edge of the next or prior range. The first range can be open ended to negative infinity, and the last range can be open ended to positive infinity.
default StringConstant: The default value if not captured by any range. If not specified, points will be filtered.

VariableValue#

Set type to the name of one of the other fields, and set only that field.

double number
computeNode ComputeNodeWithContext
duration run.api.Duration
integer integer
channel ChannelSeries deprecated: Deprecated: use series instead.
derived DerivedSeries
series SeriesVariable
string string
stringSet array of string
timestamp Timestamp

Window#

Set type to the name of one of the other fields, and set only that field.

duration DurationConstant

WindowSymmetry#

DFT_EVEN: Generates dft-even/periodic window values.
SYMMETRIC: Generates symmetric window values.

WithSeriesDataset#

Adds one or more derived series to the base dataset. Each entry’s definition must reference only the wrapping input. Series names must be unique within the request; a derived name replaces a same-named series from input.

input Dataset required: The base dataset to which the new series will be added.
series array of NamedSeries: The named derived series to add.

ZscoreSeries#

Computes the z-score for each series within a group-by family. For each group at each timestamp, the output is (value - mean) / stddev, where mean and stddev are computed across all groups in the family at that timestamp, using the interpolation configuration to align values across groups. When stddev is zero at a timestamp (e.g., all groups have the same value), that data point is excluded from the output. If groupByTags is empty, all groups in the input series will be considered as a single family. This node only produces meaningful results when the input has a group by.

input NumericSeries required
groupByTags array of StringConstant: Tags to group by for the normalization. If left empty, all groups in the input will be considered as a single family. Otherwise, inputs will be normalized separately across the groups specified by the tags.
interpolationConfiguration InterpolationConfiguration deprecated: Configuration for aligning values across groups at each timestamp. Defaults to forward fill interpolation with a 1s interpolation radius. Deprecated: Setting interpolation on this node is deprecated.

Page design based on Scalar, MIT License.

Compute service#

Endpoints#

Compute#

Parameterized compute#

Compute units#

Batch compute with units#

Batch compute units#

Compute with units#

Batch kill requests#

Batch get request progress#

Get quota usage#

Errors#

Types#

Abs#

AbsoluteThreshold#

AbsoluteTimestampSeries#

Acos#

Add#

AfterPersistenceWindow#

AggregateEnumSeries#

AggregateNumericSeries#

AggregateUnderRangesSeries#

AggregationBuilder#

Alignment#

AlignmentConfiguration#

AlignmentDriverSeries#

AllowNegativeValues#

AnchorAlignedDataset#

AnchorReference#

AndSeries#

ApproximateThresholdOperator#

ApproximateThresholdRanges#

ArithmeticSeries#

ArraySeries#

ArrowArrayPlot#

ArrowPlot#

Asin#

Asset#

AssetChannel#

AssetsSearchTarget#

Atan2#

Average#

BandPassConfiguration#

BandStopConfiguration#

BinaryArithmeticOperation#

BinaryArithmeticSeries#

BitAnd#

BitAndFunction#

BitOperationFunction#

BitOperationSeries#

BitOr#

BitOrFunction#

BitShiftLeft#

BitShiftLeftFunction#

BitShiftRight#

BitShiftRightFunction#

BitTest#

BitTestFunction#

BitXor#

BitXorFunction#

Bode#

BooleanRanges#

BooleanSeries#

BucketedCartesian3dPlot#

BucketedCartesianPlot#

BucketedEnumArrayPlot#

BucketedEnumPlot#

BucketedFrequencyDomainPlot#

BucketedMultivariatePlot#

BucketedNumericArrayPlot#

BucketedNumericPlot#

BuiltinConstant#

BuiltInCurveFit#

Cartesian#

Cartesian3d#

Cartesian3dBounds#

Cartesian3dBucket#

Cartesian3dUnitResult#

CartesianBounds#

CartesianBucket#