[CORE] optimize driver heap memory by lazily store metadata columns and fix mismatch result when input_file_name is only in filter condition#11899
Merged
Conversation
taiyang-li
changed the title
|
Run Gluten Clickhouse CI on x86 |
Contributor
Author
|
@baibaichen @rui-mo could you help review this pr, thanks! |
|
This PR is stale because it has been open 45 days with no activity. Remove stale label or comment or this will be closed in 10 days. |
|
This PR was auto-closed because it has been stalled for 10 days with no activity. Please feel free to reopen if it is still valid. Thanks. |
|
Run Gluten Clickhouse CI on x86 |
Contributor
|
any process about this pr @taiyang-li |
Contributor
Author
|
@zzcclp I've been a bit busy lately and don't have the energy to read this PR for now. I'll check it out when I have idle time |
taiyang-li
force-pushed
the
avoid_driver_oom
branch
from
60340bb to
3c42cd6
Compare
|
Run Gluten Clickhouse CI on x86 |
1 similar comment
|
Run Gluten Clickhouse CI on x86 |
taiyang-li
force-pushed
the
avoid_driver_oom
branch
from
d46120b to
ecde2a3
Compare
|
Run Gluten Clickhouse CI on x86 |
taiyang-li
changed the title
taiyang-li
changed the title
|
Run Gluten Clickhouse CI on x86 |
taiyang-li
changed the title
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changes are proposed in this pull request?
When scanning tens of millions of files, the driver may run out of heap memory during split construction.
Currently we always generate per-file metadata maps (InputFileName / InputFileBlockStart / InputFileBlockLength) for each split, even if the query does not reference these metadata columns. This causes a large number of
HashMapinstances to be kept in memory and may trigger OOM in the planner.How was this patch tested?
New ut.
Was this patch authored or co-authored using generative AI tooling?
Generated-by: trae