[SPARK-48258][PYTHON][CONNECT][FOLLOW-UP] Bind relation ID to the plan instead of DataFrame #46694
Conversation
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
HyukjinKwon
force-pushed
the
SPARK-48258-followup
branch
from
13a8d75 to
d8d5a47
Compare
Member
Author
HyukjinKwon
force-pushed
the
SPARK-48258-followup
branch
from
d8d5a47 to
549dae7
Compare
HyukjinKwon
force-pushed
the
SPARK-48258-followup
branch
from
549dae7 to
21f2d40
Compare
HyukjinKwon
commented
May 22, 2024
| ) | ||
|
|
||
| del df | ||
| gc.collect() |
Member
Author
There was a problem hiding this comment.
Unlike JVM, this does trigger the full GC
|
|
||
| def __del__(self) -> None: | ||
| session = self._spark_session | ||
| # If session is already closed, all cached DataFrame should be released. |
Contributor
There was a problem hiding this comment.
is it possible to release those cached dataframes in server side?
Member
Author
There was a problem hiding this comment.
I don't think so.. we can only tell when to release at the client side
Member
Author
There was a problem hiding this comment.
Without the change, we can only know if the session is disconnected, and we're already releasing all in this case.
zhengruifeng
approved these changes
May 22, 2024
Member
Author
|
Merged to master. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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 were proposed in this pull request?
This PR addresses #46683 (comment) comment within Python, by using ID at the plan instead of DataFrame itself.
Why are the changes needed?
Because the DataFrame holds the relation ID, if DataFrame B are derived from DataFrame A, and DataFrame A is garbage-collected, then the cache might not exist anymore. See the example below:
Does this PR introduce any user-facing change?
No, the main change has not been released out yet.
How was this patch tested?
Manually tested, and added a unittest.
Was this patch authored or co-authored using generative AI tooling?
No.