You are viewing a plain text version of this content. The canonical link for it is here.
Posted to gitbox@activemq.apache.org by GitBox <gi...@apache.org> on 2020/11/27 07:19:29 UTC
[GitHub] [activemq-artemis] franz1981 edited a comment on pull request #3363: ARTEMIS-3016 Reduce DuplicateIDCache memory footprint
franz1981 edited a comment on pull request #3363:
URL: https://github.com/apache/activemq-artemis/pull/3363#issuecomment-734683634
First results of allocation rate from this PR:
```
Benchmark (persist) (size) Mode Cnt Score Error Units
DuplicateIDCacheBenchmark.atomicVerify false 2000 thrpt 16 5023355.078 ± 36609.582 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm false 2000 thrpt 16 80.000 ± 0.001 B/op
DuplicateIDCacheBenchmark.atomicVerify false 20000 thrpt 16 4401455.813 ± 160390.631 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm false 20000 thrpt 16 80.000 ± 0.001 B/op
DuplicateIDCacheBenchmark.atomicVerify true 2000 thrpt 16 4734734.651 ± 82617.669 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm true 2000 thrpt 16 80.000 ± 0.001 B/op
DuplicateIDCacheBenchmark.atomicVerify true 20000 thrpt 16 3957606.687 ± 543564.163 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm true 20000 thrpt 16 80.000 ± 0.001 B/op
```
`master`:
```
Benchmark (persist) (size) Mode Cnt Score Error Units
DuplicateIDCacheBenchmark.atomicVerify false 2000 thrpt 16 5007048.617 ± 51566.603 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm false 2000 thrpt 16 94.976 ± 0.001 B/op
DuplicateIDCacheBenchmark.atomicVerify false 20000 thrpt 16 4345437.808 ± 160914.465 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm false 20000 thrpt 16 95.898 ± 0.001 B/op
DuplicateIDCacheBenchmark.atomicVerify true 2000 thrpt 16 4649936.243 ± 48984.066 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm true 2000 thrpt 16 118.976 ± 0.001 B/op
DuplicateIDCacheBenchmark.atomicVerify true 20000 thrpt 16 4035259.269 ± 140145.422 ops/s
DuplicateIDCacheBenchmark.atomicVerify:·gc.alloc.rate.norm true 20000 thrpt 16 119.898 ± 0.001 B/op
```
The allocation rate is a bit lower (we don't allocate anymore for `Long` and `Integer` and the memory footprint (see previous comments) is reduced, while the performance stay almost the same.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org