-
Notifications
You must be signed in to change notification settings - Fork 4k
GH-14937: [C++] String Sort / Rank Benchmarks #34811
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
Conversation
4cf771b to
6d85624
Compare
7722db1 to
2caf530
Compare
|
Python issue in CI is unrelated. |
wgtmac
left a comment
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.
Could you please provide some results? Just curious about the
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.
Is n_chunks worth being a parameter instead of a constant value?
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.
The other benchmarks also chunk the arrays by 10, so to keeps things comparable, we should use 10 everywhere.
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.
I am not familiar with the sort function here. Does it sort null values to the first or to the end? Does the null distribution affect the benchmark result? It would be good to provide some benchmark result as well.
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.
I pasted below. Null in variable-size binary arrays will reduce the amout of data in the data buffer, making CPU caching more effective.
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.
Given that you went to the trouble of making this configurable, should we have a ArraySortIndicesStringShort and ArraySortIndicesStringLong similar to how we have Narrow and Wide versions of the integer benchmarks?
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.
|
@wgtmac here it is. Number might contain noise as I run a lot of stuff on my dev machine: |
2caf530 to
f02c06d
Compare
zeroshade
left a comment
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.
this all LGTM, i think the other comments already addressed question i would have asked.
|
@zeroshade I added the Narrow and Wide versions based on @westonpace's request above. I picked 16 chars for narrow and 64 chars for the wide. The CI failure is unrelated, so feel free to merge if the last commit is OK. |
|
Thanks @felipecrv for seeing this through |
|
Benchmark runs are scheduled for baseline = 8c91434 and contender = a09201b. a09201b is a master commit associated with this PR. Results will be available as each benchmark for each run completes. |
|
['Python', 'R'] benchmarks have high level of regressions. |
This is apache#15041 with the last review comments addressed. * Closes: apache#14937 Lead-authored-by: Felipe Oliveira Carvalho <felipekde@gmail.com> Co-authored-by: Will Ayd <william.ayd@icloud.com> Signed-off-by: Matt Topol <zotthewizard@gmail.com>
This is apache#15041 with the last review comments addressed. * Closes: apache#14937 Lead-authored-by: Felipe Oliveira Carvalho <felipekde@gmail.com> Co-authored-by: Will Ayd <william.ayd@icloud.com> Signed-off-by: Matt Topol <zotthewizard@gmail.com>
This is #15041 with the last review comments addressed.