Mimir Query Engine: Add instant-vector functions

CHANGELOG.md

@@ -19,7 +19,7 @@
 * [FEATURE] Continuous-test: now runable as a module with `mimir -target=continuous-test`. #7747
 * [FEATURE] Store-gateway: Allow specific tenants to be enabled or disabled via `-store-gateway.enabled-tenants` or `-store-gateway.disabled-tenants` CLI flags or their corresponding YAML settings. #7653
 * [FEATURE] New `-<prefix>.s3.bucket-lookup-type` flag configures lookup style type, used to access bucket in s3 compatible providers. #7684
-* [FEATURE] Querier: add experimental streaming PromQL engine, enabled with `-querier.promql-engine=mimir`. #7693 #7898 #7899 #8023 #8058 #8096 #8121 #8197 #8230 #8247 #8270 #8276 #8277 #8291 #8303 #8340
+* [FEATURE] Querier: add experimental streaming PromQL engine, enabled with `-querier.promql-engine=mimir`. #7693 #7898 #7899 #8023 #8058 #8096 #8121 #8197 #8230 #8247 #8270 #8276 #8277 #8291 #8303 #8340 #8256
 * [FEATURE] New `/ingester/unregister-on-shutdown` HTTP endpoint allows dynamic access to ingesters' `-ingester.ring.unregister-on-shutdown` configuration. #7739
 * [FEATURE] Server: added experimental [PROXY protocol support](https://www.haproxy.org/download/2.3/doc/proxy-protocol.txt). The PROXY protocol support can be enabled via `-server.proxy-protocol-enabled=true`. When enabled, the support is added both to HTTP and gRPC listening ports. #7698
 * [FEATURE] mimirtool: Add `runtime-config verify` sub-command, for verifying Mimir runtime config files. #8123

pkg/streamingpromql/functions.go

@@ -0,0 +1,113 @@
+// SPDX-License-Identifier: AGPL-3.0-only
+
+package streamingpromql
+
+import (
+	"fmt"
+
+	"github.com/grafana/mimir/pkg/streamingpromql/functions"
+	"github.com/grafana/mimir/pkg/streamingpromql/operators"
+	"github.com/grafana/mimir/pkg/streamingpromql/pooling"
+	"github.com/grafana/mimir/pkg/streamingpromql/types"
+)
+
+type InstantVectorFunctionOperatorFactory func(args []types.Operator, pool *pooling.LimitingPool) (types.InstantVectorOperator, error)
+
+// SingleInputVectorFunctionOperator creates an InstantVectorFunctionOperator for functions
+// that have exactly 1 argument (v instant-vector).
+//
+// Parameters:
+//   - name: The name of the function.
+//   - metadataFunc: The function for handling metadata
+//   - seriesDataFunc: The function to handle series data
+//
+// Returns:
+//
+//	An InstantVectorFunctionOperator.
+func SingleInputVectorFunctionOperator(name string, metadataFunc functions.SeriesMetadataFunction, seriesDataFunc functions.InstantVectorFunction) InstantVectorFunctionOperatorFactory {
+	return func(args []types.Operator, pool *pooling.LimitingPool) (types.InstantVectorOperator, error) {
+		if len(args) != 1 {
+			// Should be caught by the PromQL parser, but we check here for safety.
+			return nil, fmt.Errorf("expected exactly 1 argument for %s, got %v", name, len(args))
+		}
+
+		inner, ok := args[0].(types.InstantVectorOperator)
+		if !ok {
+			// Should be caught by the PromQL parser, but we check here for safety.
+			return nil, fmt.Errorf("expected an instant vector argument for %s, got %T", name, args[0])
+		}
+
+		return &operators.FunctionOverInstantVector{
+			Inner: inner,
+			Pool:  pool,
+
+			MetadataFunc:   metadataFunc,
+			SeriesDataFunc: seriesDataFunc,
+		}, nil
+	}
+}
+
+// TransformationFunctionOperator creates an InstantVectorFunctionOperator for functions
+// that have exactly 1 argument (v instant-vector), and drop the series __name__ label.
+//
+// Parameters:
+//   - name: The name of the function.
+//   - seriesDataFunc: The function to handle series data
+//
+// Returns:
+//
+//	An InstantVectorFunctionOperator.
+func TransformationFunctionOperator(name string, seriesDataFunc functions.InstantVectorFunction) InstantVectorFunctionOperatorFactory {
+	return SingleInputVectorFunctionOperator(name, functions.DropSeriesName, seriesDataFunc)
+}
+
+// LabelManipulationFunctionOperator creates an InstantVectorFunctionOperator for functions
+// that have exactly 1 argument (v instant-vector), and need to manipulate the labels of
+// each series without manipulating the returned samples.
+// The values of v are passed through.
+//
+// Parameters:
+//   - name: The name of the function.
+//   - metadataFunc: The function for handling metadata
+//
+// Returns:
+//
+//	An InstantVectorFunctionOperator.
+func LabelManipulationFunctionOperator(name string, metadataFunc functions.SeriesMetadataFunction) InstantVectorFunctionOperatorFactory {
+	return SingleInputVectorFunctionOperator(name, metadataFunc, functions.Passthrough)
+}
+
+func rateFunctionOperator(args []types.Operator, pool *pooling.LimitingPool) (types.InstantVectorOperator, error) {
+	if len(args) != 1 {
+		// Should be caught by the PromQL parser, but we check here for safety.
+		return nil, fmt.Errorf("expected exactly 1 argument for rate, got %v", len(args))
+	}
+
+	inner, ok := args[0].(types.RangeVectorOperator)
+	if !ok {
+		// Should be caught by the PromQL parser, but we check here for safety.
+		return nil, fmt.Errorf("expected a range vector argument for rate, got %T", args[0])
+	}
+
+	return &operators.FunctionOverRangeVector{
+		Inner: inner,
+		Pool:  pool,
+	}, nil
+}
+
+// These functions return an instant-vector.
+var instantVectorFunctionOperatorFactories = map[string]InstantVectorFunctionOperatorFactory{
+	"acos":            TransformationFunctionOperator("acos", functions.Acos),
+	"histogram_count": TransformationFunctionOperator("histogram_count", functions.HistogramCount),
+	"histogram_sum":   TransformationFunctionOperator("histogram_sum", functions.HistogramSum),
+	"rate":            rateFunctionOperator,
+}
+
+func RegisterInstantVectorFunctionOperator(functionName string, functionOperator InstantVectorFunctionOperatorFactory) error {
+	if _, exists := instantVectorFunctionOperatorFactories[functionName]; exists {
+		return fmt.Errorf("function '%s' has already been registered", functionName)
+	}
+
+	instantVectorFunctionOperatorFactories[functionName] = functionOperator
+	return nil
+}

pkg/streamingpromql/functions/common.go

@@ -0,0 +1,45 @@
+// SPDX-License-Identifier: AGPL-3.0-only
+
+package functions
+
+import (
+	"github.com/grafana/mimir/pkg/streamingpromql/pooling"
+	"github.com/grafana/mimir/pkg/streamingpromql/types"
+)
+
+type SeriesMetadataFunction func(seriesMetadata []types.SeriesMetadata, pool *pooling.LimitingPool) ([]types.SeriesMetadata, error)
+
+func DropSeriesName(seriesMetadata []types.SeriesMetadata, _ *pooling.LimitingPool) ([]types.SeriesMetadata, error) {
+	for i := range seriesMetadata {
+		seriesMetadata[i].Labels = seriesMetadata[i].Labels.DropMetricName()
+	}
+
+	return seriesMetadata, nil
+}
+
+type InstantVectorFunction func(seriesData types.InstantVectorSeriesData, pool *pooling.LimitingPool) (types.InstantVectorSeriesData, error)
+
+// floatTransformationFunc is not needed elsewhere, so it is not exported yet
+func floatTransformationFunc(transform func(f float64) float64) InstantVectorFunction {
+	return func(seriesData types.InstantVectorSeriesData, pool *pooling.LimitingPool) (types.InstantVectorSeriesData, error) {
+		for i := range seriesData.Floats {
+			seriesData.Floats[i].F = transform(seriesData.Floats[i].F)
+		}
+		return seriesData, nil
+	}
+}
+
+func FloatTransformationDropHistogramsFunc(transform func(f float64) float64) InstantVectorFunction {
+	ft := floatTransformationFunc(transform)
+	return func(seriesData types.InstantVectorSeriesData, pool *pooling.LimitingPool) (types.InstantVectorSeriesData, error) {
+		// Functions that do not explicitly mention native histograms in their documentation will ignore histogram samples.
+		// https://prometheus.io/docs/prometheus/latest/querying/functions
+		pool.PutHPointSlice(seriesData.Histograms)
+		seriesData.Histograms = nil
+		return ft(seriesData, pool)
+	}
+}
+
+func Passthrough(seriesData types.InstantVectorSeriesData, _ *pooling.LimitingPool) (types.InstantVectorSeriesData, error) {
+	return seriesData, nil
+}

pkg/streamingpromql/functions/common_test.go

@@ -0,0 +1,87 @@
+// SPDX-License-Identifier: AGPL-3.0-only
+
+package functions
+
+import (
+	"testing"
+
+	"github.com/prometheus/prometheus/model/histogram"
+	"github.com/prometheus/prometheus/model/labels"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/stretchr/testify/require"
+
+	"github.com/grafana/mimir/pkg/streamingpromql/pooling"
+	"github.com/grafana/mimir/pkg/streamingpromql/types"
+)
+
+func TestDropSeriesName(t *testing.T) {
+	seriesMetadata := []types.SeriesMetadata{
+		{Labels: labels.FromStrings("__name__", "metric_name", "label1", "value1")},
+		{Labels: labels.FromStrings("__name__", "another_metric", "label2", "value2")},
+	}
+
+	expected := []types.SeriesMetadata{
+		{Labels: labels.FromStrings("label1", "value1")},
+		{Labels: labels.FromStrings("label2", "value2")},
+	}
+
+	modifiedMetadata, err := DropSeriesName(seriesMetadata, pooling.NewLimitingPool(0, nil))
+	require.NoError(t, err)
+	require.Equal(t, expected, modifiedMetadata)
+}
+
+func TestFloatTransformationFunc(t *testing.T) {
+	transform := func(f float64) float64 { return f * 2 }
+	transformFunc := floatTransformationFunc(transform)
+
+	seriesData := types.InstantVectorSeriesData{
+		Floats: []promql.FPoint{
+			{F: 1.0},
+			{F: 2.5},
+		},
+		Histograms: []promql.HPoint{
+			{H: &histogram.FloatHistogram{Count: 1, Sum: 2}},
+		},
+	}
+
+	expected := types.InstantVectorSeriesData{
+		Floats: []promql.FPoint{
+			{F: 2.0},
+			{F: 5.0},
+		},
+		Histograms: []promql.HPoint{
+			{H: &histogram.FloatHistogram{Count: 1, Sum: 2}},
+		},
+	}
+
+	modifiedSeriesData, err := transformFunc(seriesData, pooling.NewLimitingPool(0, nil))
+	require.NoError(t, err)
+	require.Equal(t, expected, modifiedSeriesData)
+}
+
+func TestFloatTransformationDropHistogramsFunc(t *testing.T) {
+	transform := func(f float64) float64 { return f * 2 }
+	transformFunc := FloatTransformationDropHistogramsFunc(transform)
+
+	seriesData := types.InstantVectorSeriesData{
+		Floats: []promql.FPoint{
+			{F: 1.0},
+			{F: 2.5},
+		},
+		Histograms: []promql.HPoint{
+			{H: &histogram.FloatHistogram{Count: 1, Sum: 2}},
+		},
+	}
+
+	expected := types.InstantVectorSeriesData{
+		Floats: []promql.FPoint{
+			{F: 2.0},
+			{F: 5.0},
+		},
+		Histograms: nil, // Histograms should be dropped
+	}
+
+	modifiedSeriesData, err := transformFunc(seriesData, pooling.NewLimitingPool(0, nil))
+	require.NoError(t, err)
+	require.Equal(t, expected, modifiedSeriesData)
+}

pkg/streamingpromql/functions/math.go

@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: AGPL-3.0-only
+
+package functions
+
+import (
+	"math"
+)
+
+var Acos = FloatTransformationDropHistogramsFunc(math.Acos)

pkg/streamingpromql/functions/math_test.go

@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: AGPL-3.0-only
+
+package functions
+
+import (
+	"math"
+	"testing"
+
+	"github.com/prometheus/prometheus/model/histogram"
+	"github.com/prometheus/prometheus/promql"
+	"github.com/stretchr/testify/require"
+
+	"github.com/grafana/mimir/pkg/streamingpromql/pooling"
+	"github.com/grafana/mimir/pkg/streamingpromql/types"
+)
+
+func TestAcos(t *testing.T) {
+	pool := pooling.NewLimitingPool(0, nil)
+
+	seriesData := types.InstantVectorSeriesData{
+		Floats: []promql.FPoint{
+			{F: 1.0},
+			{F: 0.5},
+			{F: 0.0},
+		},
+		Histograms: []promql.HPoint{
+			{H: &histogram.FloatHistogram{Count: 1, Sum: 2}},
+		},
+	}
+
+	expected := types.InstantVectorSeriesData{
+		Floats: []promql.FPoint{
+			{F: math.Acos(1.0)},
+			{F: math.Acos(0.5)},
+			{F: math.Acos(0.0)},
+		},
+		Histograms: nil, // Histograms should be dropped
+	}
+
+	modifiedSeriesData, err := Acos(seriesData, pool)
+	require.NoError(t, err)
+	require.Equal(t, expected, modifiedSeriesData)
+}

pkg/streamingpromql/functions/native_histograms.go

@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: AGPL-3.0-only
+
+package functions
+
+import (
+	"github.com/prometheus/prometheus/promql"
+
+	"github.com/grafana/mimir/pkg/streamingpromql/pooling"
+	"github.com/grafana/mimir/pkg/streamingpromql/types"
+)
+
+func HistogramCount(seriesData types.InstantVectorSeriesData, pool *pooling.LimitingPool) (types.InstantVectorSeriesData, error) {
+	floats, err := pool.GetFPointSlice(len(seriesData.Histograms))
+	if err != nil {
+		return types.InstantVectorSeriesData{}, err
+	}
+
+	data := types.InstantVectorSeriesData{
+		Floats: floats,
+	}
+	for _, histogram := range seriesData.Histograms {
+		data.Floats = append(data.Floats, promql.FPoint{
+			T: histogram.T,
+			F: histogram.H.Count,
+		})
+	}
+	pool.PutInstantVectorSeriesData(seriesData)
+	return data, nil
+}
+
+func HistogramSum(seriesData types.InstantVectorSeriesData, pool *pooling.LimitingPool) (types.InstantVectorSeriesData, error) {
+	floats, err := pool.GetFPointSlice(len(seriesData.Histograms))
+	if err != nil {
+		return types.InstantVectorSeriesData{}, err
+	}
+
+	data := types.InstantVectorSeriesData{
+		Floats: floats,
+	}
+	for _, histogram := range seriesData.Histograms {
+		data.Floats = append(data.Floats, promql.FPoint{
+			T: histogram.T,
+			F: histogram.H.Sum,
+		})
+	}
+	pool.PutInstantVectorSeriesData(seriesData)
+	return data, nil
+}