Enable AO/AOCO insert to multiple files even enable_parallel is off #83
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
avamingli
force-pushed
the
fix_ao_segfiles
branch
from
68242da
to
b69d885
Compare
|
The hack way |
my-ship-it
approved these changes
Jul 26, 2023
my-ship-it
approved these changes
Jul 27, 2023
avamingli
force-pushed
the
fix_ao_segfiles
branch
2 times, most recently
from
a662ecf
to
39fdb57
Compare
Since commit 5acbf4e, when enable_parallel is off, we will insert into only one AO segfile even gp_appendonly_insert_files is > 1. It will make data skew after user set enable_parallel to off, while there are a lot of data inserted later or there are online-steaming ETL severs. And that make our parallel plan has a bottleneck. In general, we should try to make AO segfiles as much as GUC gp_appendonly_insert_files and avoid data skew for users, no matter users use parallel query or not. And only keep gp_appendonly_insert_files default value to 4 is enough. Remove enable_parallel when initing AO/AOCS insertDesc. Authored-by: Zhang Mingli avamingli@gmail.com
Utility mode should always use segfile 0 for AO/AOCS tables. Insert multiple files in Utility mode will cause isolation2 fail when there are concurrent insertion like 0U to launch processes. Authored-by: Zhang Mingli avamingli@gmail.com
We have changed gp_appendonly_insert_files to 4 by default. It will cause lots of cases fail in cases related to AO/AOCS segfile and etc. As we enable insert multiple files regardless of enable_parallel, the problems come up. Use gpconfig to set GUCs before test cases to disable insert multiple files and etc. Authored-by: Zhang Mingli avamingli@gmail.com
avamingli
force-pushed
the
fix_ao_segfiles
branch
from
39fdb57
to
c088a88
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Since commit 5acbf4e, when enable_parallel is off, we will insert
into only one AO segfile even gp_appendonly_insert_files is > 1.
It will make data skew after user set enable_parallel to off,
while there are a lot of data inserted later or there are
online-steaming ETL severs.
And that make our parallel plan has a bottleneck.
In general, we should try to make AO segfiles as much as GUC
gp_appendonly_insert_files and avoid data skew for users,
no matter users use parallel query or not.
And only keep gp_appendonly_insert_files default value to 4 is enough.
Remove enable_parallel when initing AO/AOCS insertDesc.
Authored-by: Zhang Mingli avamingli@gmail.com
closes: #39
Change logs
Describe your change clearly, including what problem is being solved or what feature is being added.
If it has some breaking backward or forward compatibility, please clary.
Why are the changes needed?
Describe why the changes are necessary.
Does this PR introduce any user-facing change?
If yes, please clarify the previous behavior and the change this PR proposes.
How was this patch tested?
Please detail how the changes were tested, including manual tests and any relevant unit or integration tests.
Contributor's Checklist
Here are some reminders before you submit the pull request:
make installcheckmake -C src/test installcheck-cbdb-parallel