Optimise relabeling by re-using memory (#11147)

* model/relabel: Add benchmark

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* model/relabel: re-use Builder across relabels

Saves memory allocations.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* labels.Builder: allow re-use of result slice

This reduces memory allocations where the caller has a suitable slice available.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* model/relabel: re-use source values slice

To reduce memory allocations.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* Unwind one change causing test failures

Restore original behaviour in PopulateLabels, where we must not overwrite the input set.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* relabel: simplify values optimisation

Use a stack-based array for up to 16 source labels, which will be the
vast majority of cases.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* lint

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

cmd/promtool/rules.go

@@ -165,7 +165,7 @@ func (importer *ruleImporter) importRule(ctx context.Context, ruleExpr, ruleName
 				lb.Set(labels.MetricName, ruleName)
 
 				for _, value := range sample.Values {
-					if err := app.add(ctx, lb.Labels(), timestamp.FromTime(value.Timestamp.Time()), float64(value.Value)); err != nil {
+					if err := app.add(ctx, lb.Labels(nil), timestamp.FromTime(value.Timestamp.Time()), float64(value.Value)); err != nil {
 						return fmt.Errorf("add: %w", err)
 					}
 				}

model/labels/labels.go

@@ -467,17 +467,25 @@ func (b *Builder) Set(n, v string) *Builder {
 	return b
 }
 
-// Labels returns the labels from the builder. If no modifications
+// Labels returns the labels from the builder, adding them to res if non-nil.
-// were made, the original labels are returned.
+// Argument res can be the same as b.base, if caller wants to overwrite that slice.
-func (b *Builder) Labels() Labels {
+// If no modifications were made, the original labels are returned.
+func (b *Builder) Labels(res Labels) Labels {
 	if len(b.del) == 0 && len(b.add) == 0 {
 		return b.base
 	}
 
+	if res == nil {
 		// In the general case, labels are removed, modified or moved
 		// rather than added.
-	res := make(Labels, 0, len(b.base))
+		res = make(Labels, 0, len(b.base))
+	} else {
+		res = res[:0]
+	}
 Outer:
+	// Justification that res can be the same slice as base: in this loop
+	// we move forward through base, and either skip an element or assign
+	// it to res at its current position or an earlier position.
 	for _, l := range b.base {
 		for _, n := range b.del {
 			if l.Name == n {

model/labels/labels_test.go

@@ -722,7 +722,7 @@ func TestBuilder(t *testing.T) {
 				b.Keep(tcase.keep...)
 			}
 			b.Del(tcase.del...)
-			require.Equal(t, tcase.want, b.Labels())
+			require.Equal(t, tcase.want, b.Labels(tcase.base))
 		})
 	}
 }

model/relabel/relabel.go

@@ -192,24 +192,29 @@ func (re Regexp) String() string {
 // are applied in order of input.
 // If a label set is dropped, nil is returned.
 // May return the input labelSet modified.
-func Process(labels labels.Labels, cfgs ...*Config) labels.Labels {
+func Process(lbls labels.Labels, cfgs ...*Config) labels.Labels {
+	lb := labels.NewBuilder(nil)
 	for _, cfg := range cfgs {
-		labels = relabel(labels, cfg)
+		lbls = relabel(lbls, cfg, lb)
-		if labels == nil {
+		if lbls == nil {
 			return nil
 		}
 	}
-	return labels
+	return lbls
 }
 
-func relabel(lset labels.Labels, cfg *Config) labels.Labels {
+func relabel(lset labels.Labels, cfg *Config, lb *labels.Builder) labels.Labels {
-	values := make([]string, 0, len(cfg.SourceLabels))
+	var va [16]string
+	values := va[:0]
+	if len(cfg.SourceLabels) > cap(values) {
+		values = make([]string, 0, len(cfg.SourceLabels))
+	}
 	for _, ln := range cfg.SourceLabels {
 		values = append(values, lset.Get(string(ln)))
 	}
 	val := strings.Join(values, cfg.Separator)
 
-	lb := labels.NewBuilder(lset)
+	lb.Reset(lset)
 
 	switch cfg.Action {
 	case Drop:
@@ -267,7 +272,7 @@ func relabel(lset labels.Labels, cfg *Config) labels.Labels {
 		panic(fmt.Errorf("relabel: unknown relabel action type %q", cfg.Action))
 	}
 
-	return lb.Labels()
+	return lb.Labels(lset)
 }
 
 // sum64 sums the md5 hash to an uint64.

model/relabel/relabel_test.go

@@ -18,6 +18,7 @@ import (
 
 	"github.com/prometheus/common/model"
 	"github.com/stretchr/testify/require"
+	"gopkg.in/yaml.v2"
 
 	"github.com/prometheus/prometheus/model/labels"
 )
@@ -500,3 +501,160 @@ func TestTargetLabelValidity(t *testing.T) {
 			"Expected %q to be %v", test.str, test.valid)
 	}
 }
+
+func BenchmarkRelabel(b *testing.B) {
+	tests := []struct {
+		name   string
+		lbls   labels.Labels
+		config string
+		cfgs   []*Config
+	}{
+		{
+			name: "example", // From prometheus/config/testdata/conf.good.yml.
+			config: `
+            - source_labels: [job, __meta_dns_name]
+              regex: "(.*)some-[regex]"
+              target_label: job
+              replacement: foo-${1}
+              # action defaults to 'replace'
+            - source_labels: [abc]
+              target_label: cde
+            - replacement: static
+              target_label: abc
+            - regex:
+              replacement: static
+              target_label: abc`,
+			lbls: labels.FromStrings("__meta_dns_name", "example-some-x.com", "abc", "def", "job", "foo"),
+		},
+		{
+			name: "kubernetes",
+			config: `
+        - source_labels:
+            - __meta_kubernetes_pod_container_port_name
+          regex: .*-metrics
+          action: keep
+        - source_labels:
+            - __meta_kubernetes_pod_label_name
+          action: drop
+          regex: ""
+        - source_labels:
+            - __meta_kubernetes_pod_phase
+          regex: Succeeded|Failed
+          action: drop
+        - source_labels:
+            - __meta_kubernetes_pod_annotation_prometheus_io_scrape
+          regex: "false"
+          action: drop
+        - source_labels:
+            - __meta_kubernetes_pod_annotation_prometheus_io_scheme
+          target_label: __scheme__
+          regex: (https?)
+          replacement: $1
+          action: replace
+        - source_labels:
+            - __meta_kubernetes_pod_annotation_prometheus_io_path
+          target_label: __metrics_path__
+          regex: (.+)
+          replacement: $1
+          action: replace
+        - source_labels:
+            - __address__
+            - __meta_kubernetes_pod_annotation_prometheus_io_port
+          target_label: __address__
+          regex: (.+?)(\:\d+)?;(\d+)
+          replacement: $1:$3
+          action: replace
+        - regex: __meta_kubernetes_pod_annotation_prometheus_io_param_(.+)
+          replacement: __param_$1
+          action: labelmap
+        - regex: __meta_kubernetes_pod_label_prometheus_io_label_(.+)
+          action: labelmap
+        - regex: __meta_kubernetes_pod_annotation_prometheus_io_label_(.+)
+          action: labelmap
+        - source_labels:
+            - __meta_kubernetes_namespace
+            - __meta_kubernetes_pod_label_name
+          separator: /
+          target_label: job
+          replacement: $1
+          action: replace
+        - source_labels:
+          - __meta_kubernetes_namespace
+          target_label: namespace
+          action: replace
+        - source_labels:
+            - __meta_kubernetes_pod_name
+          target_label: pod
+          action: replace
+        - source_labels:
+            - __meta_kubernetes_pod_container_name
+          target_label: container
+          action: replace
+        - source_labels:
+            - __meta_kubernetes_pod_name
+            - __meta_kubernetes_pod_container_name
+            - __meta_kubernetes_pod_container_port_name
+          separator: ':'
+          target_label: instance
+          action: replace
+        - target_label: cluster
+          replacement: dev-us-central-0
+        - source_labels:
+            - __meta_kubernetes_namespace
+          regex: hosted-grafana
+          action: drop
+        - source_labels:
+            - __address__
+          target_label: __tmp_hash
+          modulus: 3
+          action: hashmod
+        - source_labels:
+            - __tmp_hash
+          regex: ^0$
+          action: keep
+        - regex: __tmp_hash
+          action: labeldrop`,
+			lbls: labels.FromStrings(
+				"__address__", "10.132.183.40:80",
+				"__meta_kubernetes_namespace", "loki-boltdb-shipper",
+				"__meta_kubernetes_pod_annotation_promtail_loki_boltdb_shipper_hash", "50523b9759094a144adcec2eae0aa4ad",
+				"__meta_kubernetes_pod_annotationpresent_promtail_loki_boltdb_shipper_hash", "true",
+				"__meta_kubernetes_pod_container_init", "false",
+				"__meta_kubernetes_pod_container_name", "promtail",
+				"__meta_kubernetes_pod_container_port_name", "http-metrics",
+				"__meta_kubernetes_pod_container_port_number", "80",
+				"__meta_kubernetes_pod_container_port_protocol", "TCP",
+				"__meta_kubernetes_pod_controller_kind", "DaemonSet",
+				"__meta_kubernetes_pod_controller_name", "promtail-loki-boltdb-shipper",
+				"__meta_kubernetes_pod_host_ip", "10.128.0.178",
+				"__meta_kubernetes_pod_ip", "10.132.183.40",
+				"__meta_kubernetes_pod_label_controller_revision_hash", "555b77cd7d",
+				"__meta_kubernetes_pod_label_name", "promtail-loki-boltdb-shipper",
+				"__meta_kubernetes_pod_label_pod_template_generation", "45",
+				"__meta_kubernetes_pod_labelpresent_controller_revision_hash", "true",
+				"__meta_kubernetes_pod_labelpresent_name", "true",
+				"__meta_kubernetes_pod_labelpresent_pod_template_generation", "true",
+				"__meta_kubernetes_pod_name", "promtail-loki-boltdb-shipper-jgtr7",
+				"__meta_kubernetes_pod_node_name", "gke-dev-us-central-0-main-n2s8-2-14d53341-9hkr",
+				"__meta_kubernetes_pod_phase", "Running",
+				"__meta_kubernetes_pod_ready", "true",
+				"__meta_kubernetes_pod_uid", "4c586419-7f6c-448d-aeec-ca4fa5b05e60",
+				"__metrics_path__", "/metrics",
+				"__scheme__", "http",
+				"__scrape_interval__", "15s",
+				"__scrape_timeout__", "10s",
+				"job", "kubernetes-pods"),
+		},
+	}
+	for i := range tests {
+		err := yaml.UnmarshalStrict([]byte(tests[i].config), &tests[i].cfgs)
+		require.NoError(b, err)
+	}
+	for _, tt := range tests {
+		b.Run(tt.name, func(b *testing.B) {
+			for i := 0; i < b.N; i++ {
+				_ = Process(tt.lbls, tt.cfgs...)
+			}
+		})
+	}
+}

notifier/notifier.go

@@ -361,7 +361,7 @@ func (n *Manager) Send(alerts ...*Alert) {
 			}
 		}
 
-		a.Labels = lb.Labels()
+		a.Labels = lb.Labels(a.Labels)
 	}
 
 	alerts = n.relabelAlerts(alerts)

promql/engine.go

@@ -2116,7 +2116,7 @@ func resultMetric(lhs, rhs labels.Labels, op parser.ItemType, matching *parser.V
 		}
 	}
 
-	ret := enh.lb.Labels()
+	ret := enh.lb.Labels(nil)
 	enh.resultMetric[str] = ret
 	return ret
 }
@@ -2156,7 +2156,7 @@ func (ev *evaluator) VectorscalarBinop(op parser.ItemType, lhs Vector, rhs Scala
 }
 
 func dropMetricName(l labels.Labels) labels.Labels {
-	return labels.NewBuilder(l).Del(labels.MetricName).Labels()
+	return labels.NewBuilder(l).Del(labels.MetricName).Labels(nil)
 }
 
 // scalarBinop evaluates a binary operation between two Scalars.
@@ -2280,7 +2280,7 @@ func (ev *evaluator) aggregation(op parser.ItemType, grouping []string, without
 		if op == parser.COUNT_VALUES {
 			lb.Reset(metric)
 			lb.Set(valueLabel, strconv.FormatFloat(s.V, 'f', -1, 64))
-			metric = lb.Labels()
+			metric = lb.Labels(nil)
 
 			// We've changed the metric so we have to recompute the grouping key.
 			recomputeGroupingKey = true
@@ -2304,7 +2304,7 @@ func (ev *evaluator) aggregation(op parser.ItemType, grouping []string, without
 			} else {
 				lb.Keep(grouping...)
 			}
-			m := lb.Labels()
+			m := lb.Labels(nil)
 			newAgg := &groupedAggregation{
 				labels:     m,
 				value:      s.V,

promql/functions.go

@@ -813,7 +813,7 @@ func funcHistogramQuantile(vals []parser.Value, args parser.Expressions, enh *Ev
 		if !ok {
 			el.Metric = labels.NewBuilder(el.Metric).
 				Del(excludedLabels...).
-				Labels()
+				Labels(nil)
 
 			mb = &metricWithBuckets{el.Metric, nil}
 			enh.signatureToMetricWithBuckets[string(enh.lblBuf)] = mb
@@ -912,7 +912,7 @@ func funcLabelReplace(vals []parser.Value, args parser.Expressions, enh *EvalNod
 				if len(res) > 0 {
 					lb.Set(dst, string(res))
 				}
-				outMetric = lb.Labels()
+				outMetric = lb.Labels(nil)
 				enh.Dmn[h] = outMetric
 			}
 		}
@@ -980,7 +980,7 @@ func funcLabelJoin(vals []parser.Value, args parser.Expressions, enh *EvalNodeHe
 				lb.Set(dst, strval)
 			}
 
-			outMetric = lb.Labels()
+			outMetric = lb.Labels(nil)
 			enh.Dmn[h] = outMetric
 		}
 
@@ -1233,14 +1233,14 @@ func createLabelsForAbsentFunction(expr parser.Expr) labels.Labels {
 			continue
 		}
 		if ma.Type == labels.MatchEqual && !m.Has(ma.Name) {
-			m = labels.NewBuilder(m).Set(ma.Name, ma.Value).Labels()
+			m = labels.NewBuilder(m).Set(ma.Name, ma.Value).Labels(nil)
 		} else {
 			empty = append(empty, ma.Name)
 		}
 	}
 
 	for _, v := range empty {
-		m = labels.NewBuilder(m).Del(v).Labels()
+		m = labels.NewBuilder(m).Del(v).Labels(nil)
 	}
 	return m
 }

rules/alerting.go

@@ -226,7 +226,7 @@ func (r *AlertingRule) sample(alert *Alert, ts time.Time) promql.Sample {
 	lb.Set(alertStateLabel, alert.State.String())
 
 	s := promql.Sample{
-		Metric: lb.Labels(),
+		Metric: lb.Labels(nil),
 		Point:  promql.Point{T: timestamp.FromTime(ts), V: 1},
 	}
 	return s
@@ -244,7 +244,7 @@ func (r *AlertingRule) forStateSample(alert *Alert, ts time.Time, v float64) pro
 	lb.Set(labels.AlertName, r.name)
 
 	s := promql.Sample{
-		Metric: lb.Labels(),
+		Metric: lb.Labels(nil),
 		Point:  promql.Point{T: timestamp.FromTime(ts), V: v},
 	}
 	return s
@@ -373,7 +373,7 @@ func (r *AlertingRule) Eval(ctx context.Context, ts time.Time, query QueryFunc,
 			annotations = append(annotations, labels.Label{Name: a.Name, Value: expand(a.Value)})
 		}
 
-		lbs := lb.Labels()
+		lbs := lb.Labels(nil)
 		h := lbs.Hash()
 		resultFPs[h] = struct{}{}
 

rules/recording.go

@@ -89,7 +89,7 @@ func (rule *RecordingRule) Eval(ctx context.Context, ts time.Time, query QueryFu
 			lb.Set(l.Name, l.Value)
 		}
 
-		sample.Metric = lb.Labels()
+		sample.Metric = lb.Labels(nil)
 	}
 
 	// Check that the rule does not produce identical metrics after applying

scrape/scrape.go

@@ -676,7 +676,7 @@ func mutateSampleLabels(lset labels.Labels, target *Target, honor bool, rc []*re
 		}
 	}
 
-	res := lb.Labels()
+	res := lb.Labels(nil)
 
 	if len(rc) > 0 {
 		res = relabel.Process(res, rc...)
@@ -716,7 +716,7 @@ func mutateReportSampleLabels(lset labels.Labels, target *Target) labels.Labels
 		lb.Set(l.Name, l.Value)
 	}
 
-	return lb.Labels()
+	return lb.Labels(nil)
 }
 
 // appender returns an appender for ingested samples from the target.

scrape/target.go

@@ -374,7 +374,7 @@ func PopulateLabels(lset labels.Labels, cfg *config.ScrapeConfig, noDefaultPort
 		}
 	}
 
-	preRelabelLabels := lb.Labels()
+	preRelabelLabels := lb.Labels(nil)
 	lset = relabel.Process(preRelabelLabels, cfg.RelabelConfigs...)
 
 	// Check if the target was dropped.
@@ -472,7 +472,7 @@ func PopulateLabels(lset labels.Labels, cfg *config.ScrapeConfig, noDefaultPort
 		lb.Set(model.InstanceLabel, addr)
 	}
 
-	res = lb.Labels()
+	res = lb.Labels(nil)
 	for _, l := range res {
 		// Check label values are valid, drop the target if not.
 		if !model.LabelValue(l.Value).IsValid() {

scrape/target_test.go

@@ -129,7 +129,7 @@ func newTestTarget(targetURL string, deadline time.Duration, lbls labels.Labels)
 	lb.Set(model.AddressLabel, strings.TrimPrefix(targetURL, "http://"))
 	lb.Set(model.MetricsPathLabel, "/metrics")
 
-	return &Target{labels: lb.Labels()}
+	return &Target{labels: lb.Labels(nil)}
 }
 
 func TestNewHTTPBearerToken(t *testing.T) {