-
Notifications
You must be signed in to change notification settings - Fork 2.4k
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
Spark-3.5: Add procedure to compute partition stats #12451
base: main
Are you sure you want to change the base?
Conversation
LOG.info("Computing partition stats for {} (branch {})", table.name(), branch); | ||
PartitionStatisticsFile statisticsFile; | ||
try { | ||
statisticsFile = PartitionStatsHandler.computeAndWriteStatsFile(table, branch); |
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.
[doubt] wouldn't it be better if we do stats computation in executor, rather than just computing it in driver in a multi-threaded way ?
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.
We did benchmarks and it turns out local algorithm is more performant than distributed way.
#9437 (comment)
So, going with non-distributed way.
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.
That makes sense, thank you for sharing the benchmark, just for my understanding what was the machine configuration used for running the benchmark ?
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.
It was JMH benchmarks on my machine (macbook m2 max), Anton also did similar test and commented on that PR.
#9437 (comment)
Depends on #12450 (Hence kept in Draft).
Fixes: #10106