-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53862/
-----------------------------------------------------------
Review request for Aurora, Mehrdad Nurolahzade and Zameer Manji.
Bugs: AURORA-1802
https://issues.apache.org/jira/browse/AURORA-1802
Repository: aurora
Description
-------
The existing benchmarks were only exercising the code paths for
mismatching offers but not for successfully launched tasks. This new
benchmark fills this gap.
In addition, this commit changes the default from DBTaskStore to
MemTaskStore for all scheduling benchmarks. Without the switch all
scheduling actions will be dominated by task store operations. This
can (and has in the past) prevent the discovery of performance
regressions.
MemTaskStore (now the default):
SchedulingBenchmarks.FillClusterBenchmark.runBenchmark thrpt 10 4.912 ±
1.790 ops/s
DBTaskStore (former default):
SchedulingBenchmarks.FillClusterBenchmark.runBenchmark thrpt 10 0.418 ±
0.076 ops/s
Diffs
-----
src/jmh/java/org/apache/aurora/benchmark/SchedulingBenchmarks.java
0510b892b67972fd15788f5debd979883c29ba6d
src/jmh/java/org/apache/aurora/benchmark/Tasks.java
e548a09d94b1c6d550f9beec2b7120b64d576f20
src/jmh/java/org/apache/aurora/benchmark/ThriftApiBenchmarks.java
7497fb96a578cff64b6484c587e29ec464969e84
Diff: https://reviews.apache.org/r/53862/diff/
Testing
-------
./gradlew jmh -Pbenchmarks='SchedulingBenchmarks.FillClusterBenchmark'
Thanks,
Stephan Erb
