puller(cdc): improve benchmarks about frontier

[hicqu]

What problem does this PR solve?

Issue Number: close #xxx

What is changed and how it works?

Check List

Tests

Benchmark results:

goos: linux
goarch: amd64
pkg: github.com/pingcap/tiflow/cdc/puller/frontier
cpu: Intel(R) Xeon(R) Gold 6240 CPU @ 2.60GHz
BenchmarkSpanForwardAndFrontier/900000(region)-10(round)-16                    1        17700041191 ns/op
BenchmarkSpanForwardAndFrontier/400000(region)-10(round)-16                    1        6722656192 ns/op

Questions

Will it cause performance regression or break compatibility?

Do you need to update user documentation, design documentation or monitoring documentation?

Release note

None

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please ask for approval from hicqu. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[overvenus]

Miss for i := 0; i < b.N; i++ { ... } ?

[hicqu]

New results including random and ordered regions:

goos: linux
goarch: amd64
pkg: github.com/pingcap/tiflow/cdc/puller/frontier
cpu: Intel(R) Xeon(R) Gold 6240 CPU @ 2.60GHz
BenchmarkSpanForwardAndFrontier/random-900000(region)-10(round)-16                     1        15379538260 ns/op
BenchmarkSpanForwardAndFrontier/random-400000(region)-10(round)-16                     1        5812647147 ns/op
BenchmarkSpanForwardAndFrontier/ordered-900000(region)-10(round)-16                    1        4241320864 ns/op
BenchmarkSpanForwardAndFrontier/ordered-400000(region)-10(round)-16                    1        1996266167 ns/op

[hicqu]

cpu: Intel(R) Xeon(R) Gold 6240 CPU @ 2.60GHz
BenchmarkSpanForwardAndFrontier_random_900k_10-16             10    15985935307 ns/op
--- BENCH: BenchmarkSpanForwardAndFrontier_random_900k_10-16
    frontier_bench_test.go:82: finishes a round, time in ms: 15936
    frontier_bench_test.go:82: finishes a round, time in ms: 17908
    frontier_bench_test.go:82: finishes a round, time in ms: 15190
    frontier_bench_test.go:82: finishes a round, time in ms: 16690
    frontier_bench_test.go:82: finishes a round, time in ms: 16611
    frontier_bench_test.go:82: finishes a round, time in ms: 16530
    frontier_bench_test.go:82: finishes a round, time in ms: 14907
    frontier_bench_test.go:82: finishes a round, time in ms: 15897
    frontier_bench_test.go:82: finishes a round, time in ms: 16225
    frontier_bench_test.go:82: finishes a round, time in ms: 14238
    ... [output truncated]
BenchmarkSpanForwardAndFrontier_ordered_900k_10-16            10    4353450849 ns/op
--- BENCH: BenchmarkSpanForwardAndFrontier_ordered_900k_10-16
    frontier_bench_test.go:82: finishes a round, time in ms: 4190
    frontier_bench_test.go:82: finishes a round, time in ms: 4381
    frontier_bench_test.go:82: finishes a round, time in ms: 3836
    frontier_bench_test.go:82: finishes a round, time in ms: 4466
    frontier_bench_test.go:82: finishes a round, time in ms: 4353
    frontier_bench_test.go:82: finishes a round, time in ms: 4759
    frontier_bench_test.go:82: finishes a round, time in ms: 4296
    frontier_bench_test.go:82: finishes a round, time in ms: 4182
    frontier_bench_test.go:82: finishes a round, time in ms: 4514
    frontier_bench_test.go:82: finishes a round, time in ms: 4302
    ... [output truncated]
BenchmarkSpanForwardAndFrontier_random_400k_10-16             10    6295161930 ns/op
--- BENCH: BenchmarkSpanForwardAndFrontier_random_400k_10-16
    frontier_bench_test.go:82: finishes a round, time in ms: 5351
    frontier_bench_test.go:82: finishes a round, time in ms: 5825
    frontier_bench_test.go:82: finishes a round, time in ms: 7055
    frontier_bench_test.go:82: finishes a round, time in ms: 6295
    frontier_bench_test.go:82: finishes a round, time in ms: 6559
    frontier_bench_test.go:82: finishes a round, time in ms: 6832
    frontier_bench_test.go:82: finishes a round, time in ms: 6155
    frontier_bench_test.go:82: finishes a round, time in ms: 6012
    frontier_bench_test.go:82: finishes a round, time in ms: 6599
    frontier_bench_test.go:82: finishes a round, time in ms: 5589
    ... [output truncated]
BenchmarkSpanForwardAndFrontier_ordered_400k_10-16            10    1737459993 ns/op
--- BENCH: BenchmarkSpanForwardAndFrontier_ordered_400k_10-16
    frontier_bench_test.go:82: finishes a round, time in ms: 1903
    frontier_bench_test.go:82: finishes a round, time in ms: 1707
    frontier_bench_test.go:82: finishes a round, time in ms: 1933
    frontier_bench_test.go:82: finishes a round, time in ms: 1801
    frontier_bench_test.go:82: finishes a round, time in ms: 1453
    frontier_bench_test.go:82: finishes a round, time in ms: 1787
    frontier_bench_test.go:82: finishes a round, time in ms: 1834
    frontier_bench_test.go:82: finishes a round, time in ms: 1499
    frontier_bench_test.go:82: finishes a round, time in ms: 1958
    frontier_bench_test.go:82: finishes a round, time in ms: 1522
    ... [output truncated]

Conclusions:
when Frontier handles ResolvedTs events, the effeciency can be influenced by whether regions are contiguous or not.

[ti-chi-bot]

[FORMAT CHECKER NOTIFICATION]

Notice: To remove the do-not-merge/needs-linked-issue label, please provide the linked issue number on one line in the PR body, for example: Issue Number: close #123 or Issue Number: ref #456.

_{📖 For more info, you can check the "Contribute Code" section in the development guide.}

[ti-chi-bot]

@hicqu: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
jenkins-ticdc/verify	719751e	link	true	/test verify

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

[nongfushanquan]

/close

[ti-chi-bot]

@nongfushanquan: Closed this PR.

In response to this:

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.