Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.
Sign upsql: add merge joiner benchmark and bytes usage optimization #31216
Conversation
changangela
requested a review
from
knz
Oct 10, 2018
changangela
requested review from
cockroachdb/distsql-prs
as
code owners
Oct 10, 2018
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
changangela
changed the title from
sql: use sub-monitor for merge joiner to allocate reserved bytes
to
sql: add merge joiner benchmark and bytes usage optimization
Oct 10, 2018
knz
approved these changes
Oct 10, 2018
Can you also include the new benchmark diff in the commit message?
Reviewed 6 of 6 files at r1, 3 of 3 files at r2.
Reviewable status:complete! 0 of 0 LGTMs obtained (and 1 stale)
knz
reviewed
Oct 10, 2018
Reviewed 2 of 2 files at r3.
Reviewable status:complete! 0 of 0 LGTMs obtained (and 1 stale)
pkg/util/mon/bytes_usage.go, line 480 at r3 (raw file):
} // Empty shrinks the account to use 0 bytes, while maintaining the minAllocated
-
this comment refers to the now-inexistent minAllocated field
-
it does something fancy with
reserved, this needs to be explained.
changangela
reviewed
Oct 10, 2018
Reviewable status:
complete! 0 of 0 LGTMs obtained (and 1 stale)
pkg/util/mon/bytes_usage.go, line 480 at r3 (raw file):
Previously, knz (kena) wrote…
this comment refers to the now-inexistent minAllocated field
it does something fancy with
reserved, this needs to be explained.
Oops, fixed.
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
knz
Oct 11, 2018
Member
You write this PR is meant to "significantly improve the merge joiner performance" yet the benchmark results you pasted above do not clearly show that. What do you think?
|
You write this PR is meant to "significantly improve the merge joiner performance" yet the benchmark results you pasted above do not clearly show that. What do you think? |
knz
reviewed
Oct 11, 2018
Reviewed 1 of 1 files at r4.
Reviewable status:complete! 0 of 0 LGTMs obtained (and 1 stale)
changangela
reviewed
Oct 11, 2018
I will benchmark this change against release-2.1 instead, because master technically already has this exact optimization.
Reviewable status:
complete! 0 of 0 LGTMs obtained (and 1 stale)
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
LGTM thanks! |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
|
bors r+ |
bot
pushed a commit
that referenced
this pull request
Oct 11, 2018
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
craig
bot
commented
Oct 11, 2018
Build succeeded |
changangela commentedOct 10, 2018
•
edited
Reverted #30924 for now to compare different approaches (discussed in #31191). This PR is mainly for adding some merge joiner benchmarks as well a small change in
BoundAccount.Shrink()that significantly improves the merge joiner performance. This way, we can safely backport this change.MergeJoinerBenchmark against
release-2.1:MergeJoinerBenchmark compared with
master(masteralready has this exact optimization, we want to ensure that the performance did not deteriorate)