RandomTIsSliceSeriesEquivalent Property

public override bool IsSliceSeriesEquivalent { get; }

Public Overrides ReadOnly Property IsSliceSeriesEquivalent As Boolean
	Get

public:
virtual property bool IsSliceSeriesEquivalent {
	bool get () override;
}

abstract IsSliceSeriesEquivalent : bool with get
override IsSliceSeriesEquivalent : bool with get

function get_IsSliceSeriesEquivalent();

Property Value

Implements

When set to true, property indicates that the function, when dealing with time series data, produces the same result matrix whether processed horizontally (series-by-series) or vertically (slice-by-slice). This is typically applicable to functions that return a series of values and are designed to operate across time-aligned data series.

Setting this property to true enables certain optimizations. For instance, the system can bypass slice-mode processing and use standard series processing instead, as both approaches yield equivalent results. This can lead to performance improvements, especially in data-intensive scenarios.

It is important for users defining custom functions to accurately set this property. If the function's output does not depend on the specific mode of time series processing (series vs. slice), set this property to true. Conversely, if the function yields different results when processed by slice-by-slice as compared to when processed series-by-series, or it is a scalar function, this property should be set to false.

Incorrectly setting this property may lead to unexpected behavior or suboptimal performance, as the system relies on this flag to determine the most efficient processing approach.

Internally this function descriptor provides a hint to the Grafana data source adapter on how to handle series results for a function. For functions that return a scalar value this property is ignored, but is automatically set to false to preserve meaning. For slice operations on functions that return a series of values and produce the same result matrix whether processed horizontally or vertically, the slice operation is equivalent to its non-slice operation using the IntervalT function over the same expression, for example, the following queries are equivalent:

• SliceShift(0.02, 1, FILTER TOP 10 ActiveMeasurements WHERE SignalType='FREQ') -- and --
  Shift(1, Interval(0.02, FILTER TOP 10 ActiveMeasurements WHERE SignalType='FREQ'))
• SliceRound(0.0333, 3, ACME-STAR:FREQ; ACME-PLUS:FREQ) -- and --
  Round(3, Interval(0.0333, ACME-STAR:FREQ; ACME-PLUS:FREQ))

Reference