prometheus/rules/recording_test.go
beorn7 c0879d64cf promql: Separate Point into FPoint and HPoint
In other words: Instead of having a “polymorphous” `Point` that can
either contain a float value or a histogram value, use an `FPoint` for
floats and an `HPoint` for histograms.

This seemingly small change has a _lot_ of repercussions throughout
the codebase.

The idea here is to avoid the increase in size of `Point` arrays that
happened after native histograms had been added.

The higher-level data structures (`Sample`, `Series`, etc.) are still
“polymorphous”. The same idea could be applied to them, but at each
step the trade-offs needed to be evaluated.

The idea with this change is to do the minimum necessary to get back
to pre-histogram performance for functions that do not touch
histograms. Here are comparisons for the `changes` function. The test
data doesn't include histograms yet. Ideally, there would be no change
in the benchmark result at all.

First runtime v2.39 compared to directly prior to this commit:

```
name                                                  old time/op    new time/op    delta
RangeQuery/expr=changes(a_one[1d]),steps=1-16            391µs ± 2%     542µs ± 1%  +38.58%  (p=0.000 n=9+8)
RangeQuery/expr=changes(a_one[1d]),steps=10-16           452µs ± 2%     617µs ± 2%  +36.48%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_one[1d]),steps=100-16         1.12ms ± 1%    1.36ms ± 2%  +21.58%  (p=0.000 n=8+10)
RangeQuery/expr=changes(a_one[1d]),steps=1000-16        7.83ms ± 1%    8.94ms ± 1%  +14.21%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_ten[1d]),steps=1-16           2.98ms ± 0%    3.30ms ± 1%  +10.67%  (p=0.000 n=9+10)
RangeQuery/expr=changes(a_ten[1d]),steps=10-16          3.66ms ± 1%    4.10ms ± 1%  +11.82%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_ten[1d]),steps=100-16         10.5ms ± 0%    11.8ms ± 1%  +12.50%  (p=0.000 n=8+10)
RangeQuery/expr=changes(a_ten[1d]),steps=1000-16        77.6ms ± 1%    87.4ms ± 1%  +12.63%  (p=0.000 n=9+9)
RangeQuery/expr=changes(a_hundred[1d]),steps=1-16       30.4ms ± 2%    32.8ms ± 1%   +8.01%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=10-16      37.1ms ± 2%    40.6ms ± 2%   +9.64%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=100-16      105ms ± 1%     117ms ± 1%  +11.69%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=1000-16     783ms ± 3%     876ms ± 1%  +11.83%  (p=0.000 n=9+10)
```

And then runtime v2.39 compared to after this commit:

```
name                                                  old time/op    new time/op    delta
RangeQuery/expr=changes(a_one[1d]),steps=1-16            391µs ± 2%     547µs ± 1%  +39.84%  (p=0.000 n=9+8)
RangeQuery/expr=changes(a_one[1d]),steps=10-16           452µs ± 2%     616µs ± 2%  +36.15%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_one[1d]),steps=100-16         1.12ms ± 1%    1.26ms ± 1%  +12.20%  (p=0.000 n=8+10)
RangeQuery/expr=changes(a_one[1d]),steps=1000-16        7.83ms ± 1%    7.95ms ± 1%   +1.59%  (p=0.000 n=10+8)
RangeQuery/expr=changes(a_ten[1d]),steps=1-16           2.98ms ± 0%    3.38ms ± 2%  +13.49%  (p=0.000 n=9+10)
RangeQuery/expr=changes(a_ten[1d]),steps=10-16          3.66ms ± 1%    4.02ms ± 1%   +9.80%  (p=0.000 n=10+9)
RangeQuery/expr=changes(a_ten[1d]),steps=100-16         10.5ms ± 0%    10.8ms ± 1%   +3.08%  (p=0.000 n=8+10)
RangeQuery/expr=changes(a_ten[1d]),steps=1000-16        77.6ms ± 1%    78.1ms ± 1%   +0.58%  (p=0.035 n=9+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=1-16       30.4ms ± 2%    33.5ms ± 4%  +10.18%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=10-16      37.1ms ± 2%    40.0ms ± 1%   +7.98%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=100-16      105ms ± 1%     107ms ± 1%   +1.92%  (p=0.000 n=10+10)
RangeQuery/expr=changes(a_hundred[1d]),steps=1000-16     783ms ± 3%     775ms ± 1%   -1.02%  (p=0.019 n=9+9)
```

In summary, the runtime doesn't really improve with this change for
queries with just a few steps. For queries with many steps, this
commit essentially reinstates the old performance. This is good
because the many-step queries are the one that matter most (longest
absolute runtime).

In terms of allocations, though, this commit doesn't make a dent at
all (numbers not shown). The reason is that most of the allocations
happen in the sampleRingIterator (in the storage package), which has
to be addressed in a separate commit.

Signed-off-by: beorn7 <beorn@grafana.com>
2023-04-13 19:25:16 +02:00

262 lines
7.1 KiB
Go

// Copyright 2013 The Prometheus Authors
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package rules
import (
"context"
"testing"
"time"
"github.com/stretchr/testify/require"
"github.com/prometheus/prometheus/model/labels"
"github.com/prometheus/prometheus/model/timestamp"
"github.com/prometheus/prometheus/promql"
"github.com/prometheus/prometheus/promql/parser"
"github.com/prometheus/prometheus/util/teststorage"
)
var (
ruleEvaluationTime = time.Unix(0, 0).UTC()
exprWithMetricName, _ = parser.ParseExpr(`sort(metric)`)
exprWithoutMetricName, _ = parser.ParseExpr(`sort(metric + metric)`)
)
var ruleEvalTestScenarios = []struct {
name string
ruleLabels labels.Labels
expr parser.Expr
expected promql.Vector
}{
{
name: "no labels in recording rule, metric name in query result",
ruleLabels: labels.EmptyLabels(),
expr: exprWithMetricName,
expected: promql.Vector{
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "1", "label_b", "3"),
F: 1,
T: timestamp.FromTime(ruleEvaluationTime),
},
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "2", "label_b", "4"),
F: 10,
T: timestamp.FromTime(ruleEvaluationTime),
},
},
},
{
name: "only new labels in recording rule, metric name in query result",
ruleLabels: labels.FromStrings("extra_from_rule", "foo"),
expr: exprWithMetricName,
expected: promql.Vector{
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "1", "label_b", "3", "extra_from_rule", "foo"),
F: 1,
T: timestamp.FromTime(ruleEvaluationTime),
},
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "2", "label_b", "4", "extra_from_rule", "foo"),
F: 10,
T: timestamp.FromTime(ruleEvaluationTime),
},
},
},
{
name: "some replacement labels in recording rule, metric name in query result",
ruleLabels: labels.FromStrings("label_a", "from_rule"),
expr: exprWithMetricName,
expected: promql.Vector{
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "from_rule", "label_b", "3"),
F: 1,
T: timestamp.FromTime(ruleEvaluationTime),
},
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "from_rule", "label_b", "4"),
F: 10,
T: timestamp.FromTime(ruleEvaluationTime),
},
},
},
{
name: "no labels in recording rule, no metric name in query result",
ruleLabels: labels.EmptyLabels(),
expr: exprWithoutMetricName,
expected: promql.Vector{
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "1", "label_b", "3"),
F: 2,
T: timestamp.FromTime(ruleEvaluationTime),
},
promql.Sample{
Metric: labels.FromStrings("__name__", "test_rule", "label_a", "2", "label_b", "4"),
F: 20,
T: timestamp.FromTime(ruleEvaluationTime),
},
},
},
}
func setUpRuleEvalTest(t require.TestingT) *promql.Test {
suite, err := promql.NewTest(t, `
load 1m
metric{label_a="1",label_b="3"} 1
metric{label_a="2",label_b="4"} 10
`)
require.NoError(t, err)
return suite
}
func TestRuleEval(t *testing.T) {
suite := setUpRuleEvalTest(t)
defer suite.Close()
require.NoError(t, suite.Run())
for _, scenario := range ruleEvalTestScenarios {
t.Run(scenario.name, func(t *testing.T) {
rule := NewRecordingRule("test_rule", scenario.expr, scenario.ruleLabels)
result, err := rule.Eval(suite.Context(), ruleEvaluationTime, EngineQueryFunc(suite.QueryEngine(), suite.Storage()), nil, 0)
require.NoError(t, err)
require.Equal(t, scenario.expected, result)
})
}
}
func BenchmarkRuleEval(b *testing.B) {
suite := setUpRuleEvalTest(b)
defer suite.Close()
require.NoError(b, suite.Run())
for _, scenario := range ruleEvalTestScenarios {
b.Run(scenario.name, func(b *testing.B) {
rule := NewRecordingRule("test_rule", scenario.expr, scenario.ruleLabels)
b.ResetTimer()
for i := 0; i < b.N; i++ {
_, err := rule.Eval(suite.Context(), ruleEvaluationTime, EngineQueryFunc(suite.QueryEngine(), suite.Storage()), nil, 0)
if err != nil {
require.NoError(b, err)
}
}
})
}
}
// TestRuleEvalDuplicate tests for duplicate labels in recorded metrics, see #5529.
func TestRuleEvalDuplicate(t *testing.T) {
storage := teststorage.New(t)
defer storage.Close()
opts := promql.EngineOpts{
Logger: nil,
Reg: nil,
MaxSamples: 10,
Timeout: 10 * time.Second,
}
engine := promql.NewEngine(opts)
ctx, cancelCtx := context.WithCancel(context.Background())
defer cancelCtx()
now := time.Now()
expr, _ := parser.ParseExpr(`vector(0) or label_replace(vector(0),"test","x","","")`)
rule := NewRecordingRule("foo", expr, labels.FromStrings("test", "test"))
_, err := rule.Eval(ctx, now, EngineQueryFunc(engine, storage), nil, 0)
require.Error(t, err)
require.EqualError(t, err, "vector contains metrics with the same labelset after applying rule labels")
}
func TestRecordingRuleLimit(t *testing.T) {
suite, err := promql.NewTest(t, `
load 1m
metric{label="1"} 1
metric{label="2"} 1
`)
require.NoError(t, err)
defer suite.Close()
require.NoError(t, suite.Run())
tests := []struct {
limit int
err string
}{
{
limit: 0,
},
{
limit: -1,
},
{
limit: 2,
},
{
limit: 1,
err: "exceeded limit of 1 with 2 series",
},
}
expr, _ := parser.ParseExpr(`metric > 0`)
rule := NewRecordingRule(
"foo",
expr,
labels.FromStrings("test", "test"),
)
evalTime := time.Unix(0, 0)
for _, test := range tests {
_, err := rule.Eval(suite.Context(), evalTime, EngineQueryFunc(suite.QueryEngine(), suite.Storage()), nil, test.limit)
if err != nil {
require.EqualError(t, err, test.err)
} else if test.err != "" {
t.Errorf("Expected error %s, got none", test.err)
}
}
}
// TestRecordingEvalWithOrigin checks that the recording rule details are passed through the context.
func TestRecordingEvalWithOrigin(t *testing.T) {
ctx := context.Background()
now := time.Now()
const (
name = "my-recording-rule"
query = `count(metric{foo="bar"})`
)
var (
detail RuleDetail
lbs = labels.FromStrings("foo", "bar")
)
expr, err := parser.ParseExpr(query)
require.NoError(t, err)
rule := NewRecordingRule(name, expr, lbs)
_, err = rule.Eval(ctx, now, func(ctx context.Context, qs string, _ time.Time) (promql.Vector, error) {
detail = FromOriginContext(ctx)
return nil, nil
}, nil, 0)
require.NoError(t, err)
require.Equal(t, detail, NewRuleDetail(rule))
}