yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1321786141
> One things that I know need to be addressed are: Some merge data infos are
not saved on the driver because they are too small ( controlled by
`spark.shuffle.push.minShuffleSizeToWait`)
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1321687314
> @mridulm @wankunde @otterc I'm not sure if I missed any logic, please help
review my code , thanks~ I will improve my code style later. Now I don't change
my code in
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1321685369
@mridulm
I will speed up to finish the unfinished parts of the previous PR together
in this PR.
From your comments in the previous PR
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1316389795
@mridulm as your comment said
https://github.com/apache/spark/pull/37922#discussion_r990763769 , I want to
Improve this part of the deletion logic
--
This is an automated message from
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1311193090
my latest implementation no longer passes reduceIds from driver, there are
still some code style improvements, just some rough implementation for now
--
This is an automated message
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1311176907
@mridulm Yes...These two issues are the similar. @wankunde Can I continue
editing my PR in this Issue?
--
This is an automated message from the Apache Git Service.
To respond to the
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1310162375
> I am wondering whether the driver needs to pass the merged reduceId to the
external shuffle service (but now the driver cannot fully record merged info),
or the shuffle service records
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1308437041
I am wondering whether the driver needs to pass the merged reduceId to the
external shuffle service, or the shuffle service records the merged reduceIds,
and the subsequent driver nodes
yabola commented on PR #38560:
URL: https://github.com/apache/spark/pull/38560#issuecomment-1307984001
The two things that I know need to be addressed are:
1. Some merge data blocks are not saved on the driver because they are too
small ( controlled by