[
https://issues.apache.org/jira/browse/FLINK-15171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16993470#comment-16993470
]
Roman Grebennikov commented on FLINK-15171:
-------------------------------------------
Results of the fix:
{noformat}
1 with-fix SerializationFrameworkMiniBenchmarks.serializerTuple thrpt 50
628.377 ± 14.407 ops/ms
1 no-fix SerializationFrameworkMiniBenchmarks.serializerTuple thrpt 50
628.152 ± 11.835 ops/ms
8 with-fix SerializationFrameworkMiniBenchmarks.serializerTuple thrpt 50
451.422 ± 5.737 ops/ms
8 no-fix SerializationFrameworkMiniBenchmarks.serializerTuple thrpt 50
450.485 ± 6.710 ops/ms
16 with-fix SerializationFrameworkMiniBenchmarks.serializerTuple thrpt 50
260.823 ± 4.057 ops/ms
16 no-fix SerializationFrameworkMiniBenchmarks.serializerTuple thrpt 50
190.611 ± 2.663 ops/ms
{noformat}
TLDR: so if you stress the GC hard enough, it will start interfering with the
benchmark code.
> Performance regression in serialisation benchmarks
> --------------------------------------------------
>
> Key: FLINK-15171
> URL: https://issues.apache.org/jira/browse/FLINK-15171
> Project: Flink
> Issue Type: Bug
> Components: API / Type Serialization System, Benchmarks
> Affects Versions: 1.10.0
> Reporter: Piotr Nowojski
> Assignee: Roman Khachatryan
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 1.10.0
>
> Attachments: dec05.svg, dec11.svg
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> There is quite significant performance regression in serialisation benchmarks
> in the commit range 2ecf7ca..9320f34 (which includes FLINK-14346).
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerTuple&env=2
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerRow&env=2
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerPojo&env=2
> it coincides with the performance improvement for heavy strings
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerHeavyString&env=2
> it might be caused by some accidental change in the benchmarking code
> (changing parallelism in one benchmarks is carried on to the next one?) or in
> the code itself.
> CC [~rgrebennikov] [~AHeise]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
