-
Notifications
You must be signed in to change notification settings - Fork 888
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Account for filters in DecompressChunk row estimates #6563
base: main
Are you sure you want to change the base?
Conversation
We don't account for them at all currently, which leads to weird planning behavior, like the filters not influencing the position of relations in join.
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## main #6563 +/- ##
==========================================
- Coverage 79.73% 79.70% -0.04%
==========================================
Files 188 188
Lines 36768 36726 -42
Branches 9295 9280 -15
==========================================
- Hits 29318 29271 -47
- Misses 3112 3126 +14
+ Partials 4338 4329 -9 ☔ View full report in Codecov by Sentry. |
…ially_compressed-*
…ially_compressed-*
…ially_compressed-*
INSERT INTO mapc_test | ||
SELECT time, device, device * 0.1 | ||
FROM generate_series('2020-01-02'::timestamptz,'2020-01-18'::timestamptz,'6 hour') time, | ||
generate_series(1,3) device; | ||
FROM generate_series('2020-01-02'::timestamptz,'2020-01-15'::timestamptz,'4 second') time, | ||
generate_series(1,10) device; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
After adding the missed ANALYZE, I had to add much more rows to keep the old plans. Also renamed it to something less generic and removed the third chunk which doesn't really add coverage.
@mkindahl, @antekresic: please review this pull request.
|
We don't account for them at all currently, which leads to weird planning behavior, like the filters not influencing the position of relations in join. We don't normally have the statistics on uncompressed tables of compressed chunks, so these estimates will be using the Postgres defaults that are chosen based on the kind of filter.
Disable-check: force-changelog-file