Github user milleruntime commented on the issue:
https://github.com/apache/accumulo/pull/257
Refactored the benchmark suite to build different version. Here are the
new results:
<pre>
Test: MyBenchmark.seekAllSystemIters10 showed 89.8% increase in ops/s
Test: MyBenchmark.seekAllSystemIters1000 showed 93.8% increase in ops/s
Test: MyBenchmark.seekTestWithCQFilter10 showed -13.4% increase in ops/s
Test: MyBenchmark.seekTestWithCQFilter1000 showed 4.2% increase in ops/s
Test: MyBenchmark.seekTestWithCfFilter10 showed 35.7% increase in ops/s
Test: MyBenchmark.seekTestWithCfFilter1000 showed 63.3% increase in ops/s
Test: MyBenchmark.seekTestWithEmptyAuths10 showed 112.1% increase in ops/s
Test: MyBenchmark.seekTestWithEmptyAuths1000 showed 199.9% increase in ops/s
Total average 73.17% increase in ops/s
</pre>
The one scenario still seems to show some regression, seekTestWithCQFilter,
which seeks to data with a Column Qualifer. I am not sure why this is
happening... perhaps the extra checks added in ColumnQualifierFilter cause a
slight slowdown when a CQ is present?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
