Hello Dan Burkert, Adar Dembo,
I'd like you to do a code review. Please visit
http://gerrit.cloudera.org:8080/6908
to review the following change.
Change subject: KUDU-2017. Fix calculation of perf improvement for flushes
......................................................................
KUDU-2017. Fix calculation of perf improvement for flushes
The code to calculate the MM "perf improvement score" for a flush had two bugs:
(1) we calculated threshold - current_usage instead of current_usage -
threshold, which resulted in a negative score.
(2) we had an unsigned integer overflow, which resulted in the above negative
score becoming insanely large.
These two wrongs "made a right" in which we'd still trigger flushes at the
flush threshold, which is why this went unnoticed for quite some time. However,
the flushing behavior is more aggressive than originally intended, and we would
lose the correct prioritization of flushing tablets that are farther above the
threshold.
This patch addresses the issue.
Change-Id: I8bcf443db857ce172b8e83758eb73a24c4b0c6af
---
M src/kudu/tablet/tablet_replica-test.cc
M src/kudu/tablet/tablet_replica_mm_ops.cc
2 files changed, 6 insertions(+), 5 deletions(-)
git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/08/6908/1
--
To view, visit http://gerrit.cloudera.org:8080/6908
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: I8bcf443db857ce172b8e83758eb73a24c4b0c6af
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Dan Burkert <[email protected]>
