Use tournament loser tree for k-way sort-merging, increase merge speed by 50% by richox · Pull Request #4301 · apache/datafusion

richox · 2022-11-21T08:21:28Z

Which issue does this PR close?

Rationale for this change

What changes are included in this PR?

Are these changes tested?

running benchmarks with cargo bench --bench merge:

merge i64 time: [8.0959 ms 8.2001 ms 8.3260 ms]
change: [-58.629% -57.906% -57.138%] (p = 0.00 < 0.05)
Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
4 (4.00%) high mild
7 (7.00%) high severe
merge f64 time: [8.2829 ms 8.3378 ms 8.4028 ms]
change: [-56.127% -55.475% -54.829%] (p = 0.00 < 0.05)
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
5 (5.00%) high mild
5 (5.00%) high severe
merge utf8 low cardinality
time: [7.5418 ms 7.6425 ms 7.7538 ms]
change: [-65.897% -65.335% -64.703%] (p = 0.00 < 0.05)
Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
7 (7.00%) high mild
6 (6.00%) high severe
merge utf8 high cardinality
time: [9.9438 ms 10.094 ms 10.257 ms]
change: [-52.043% -51.155% -50.249%] (p = 0.00 < 0.05)
Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) high mild
4 (4.00%) high severe
merge utf8 tuple time: [20.197 ms 20.493 ms 20.826 ms]
change: [-40.663% -39.505% -38.279%] (p = 0.00 < 0.05)
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
6 (6.00%) high mild
4 (4.00%) high severe
merge utf8 dictionary time: [5.1537 ms 5.2135 ms 5.2788 ms]
change: [-72.062% -71.616% -71.137%] (p = 0.00 < 0.05)
Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
7 (7.00%) high mild
1 (1.00%) high severe
merge utf8 dictionary tuple
time: [7.9585 ms 8.0592 ms 8.1797 ms]
change: [-63.981% -63.324% -62.678%] (p = 0.00 < 0.05)
Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
6 (6.00%) high mild
5 (5.00%) high severe
merge mixed utf8 dictionary tuple
time: [14.528 ms 14.669 ms 14.829 ms]
change: [-43.886% -43.201% -42.489%] (p = 0.00 < 0.05)
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
5 (5.00%) high mild
5 (5.00%) high severe
merge mixed tuple time: [17.824 ms 18.068 ms 18.333 ms]
change: [-39.957% -38.888% -37.748%] (p = 0.00 < 0.05)
Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
4 (4.00%) high mild
1 (1.00%) high severe

Are there any user-facing changes?

alamb · 2022-11-21T15:11:20Z

cc @tustvold

tustvold

This looks really good, thank you, I left some comments on how to make the implementation a little easier to follow, but I would be happy for this to go in as is.

tustvold · 2022-11-21T17:26:13Z

-
- if self.in_progress.len() == self.batch_size {
- return Poll::Ready(Some(self.build_record_batch()));
+ let mut cmp_node = (num_streams + winner) / 2;


Suggested change
letmut cmp_node = (num_streams + winner) / 2;
// Replace overall winner by walking tree of losers
letmut cmp_node = (num_streams + winner) / 2;

tustvold · 2022-11-21T17:26:50Z

+ /// the loser nodes
+ loser_tree: Vec<usize>,
+
+ /// Identify whether the loser tree is adjusted


Suggested change
/// Identify whether the loser tree is adjusted
/// Identify whether the most recently yielded overall winner has been replaced
/// within the loser tree, a value of `false` indicates that they overall winner
/// has been yielded but the loser tree has not been updated
Or something to make it clearer what adjusted actually means.
FWIW a boolean of should_replace_winner or something might be clearer

tustvold · 2022-11-21T17:40:14Z

- self.current()
- .cmp(&other.current())
- .then_with(|| self.stream_idx.cmp(&other.stream_idx))
+ match (self.is_finished(), other.is_finished()) {


Suggested change
match(self.is_finished(), other.is_finished()){
// Order finished cursors last
match(self.is_finished(), other.is_finished()){

tustvold · 2022-11-21T17:45:34Z

+
+ // Init all cursors and the loser tree in the first poll
+ if self.loser_tree.is_empty() {
+ // Ensure all non-exhausted streams have a cursor from which


It might be easier to follow if this method were split into a method called init_loser_tree with a doc comment explaining what it does

tustvold · 2022-11-21T17:46:34Z

- self.cursor_finished[stream_idx] = true;
+ // Adjust the loser tree if necessary
+ if !self.loser_tree_adjusted {
+ let mut winner = self.loser_tree[0];


It might be easier to follow if this was moved into a method called replace_loser_tree_winner, perhaps with a link to this GIF - https://en.wikipedia.org/wiki/K-way_merge_algorithm#/media/File:Loser_tree_replacement_selection.gif

tustvold · 2022-11-21T17:52:05Z

- }
- }
- }
+ let min_cursor_idx = self.loser_tree[0];


I think this could be made easier to follow if it were written along the lines of
let min_cursor = self.cursors[min_cursor_idx]; if min_cursor.is_finished() { // All streams are exhausted return Poll::Ready((!self.in_progress.is_empty()).then(|| self.build_record_batch())) } self.loser_tree_adjusted = false; self.in_progress.push(...) if self.in_progress.len() == self.batch_size { return Poll::Ready(Some(self.build_record_batch())); }

I couldn't make this work in #4407

viirya · 2022-11-21T23:25:01Z

+ if challenger_win {
+ self.loser_tree[cmp_node] = winner;
+ winner = challenger;
+ } else {
+ self.loser_tree[cmp_node] = challenger;
+ }


Suggested change
if challenger_win {
self.loser_tree[cmp_node] = winner;
winner = challenger;
}else{
self.loser_tree[cmp_node] = challenger;
}
if challenger_win {
self.loser_tree[cmp_node] = winner;
winner = challenger;
}

alamb · 2022-11-22T15:18:02Z

Thanks @richox @tustvold and @viirya -- this looks awesome. Let's wait to see if @richox would like to respond to comments.

alamb · 2022-11-26T12:46:41Z

If we don't hear back from @richox by Monday I plan to merge this PR and then create a follow on PR with the proposed changes from @tustvold and @viirya

alamb · 2022-11-28T20:09:20Z

I will have a follow on PR up shortly

alamb · 2022-11-28T20:09:32Z

Thanks again @richox @tustvold and @viirya

ursabot · 2022-11-28T20:12:33Z

Benchmark runs are scheduled for baseline = 52e198e and contender = 0d334cf. 0d334cf is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on ec2-t3-xlarge-us-east-2] ec2-t3-xlarge-us-east-2
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on test-mac-arm] test-mac-arm
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on ursa-i9-9960x] ursa-i9-9960x
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on ursa-thinkcentre-m75q] ursa-thinkcentre-m75q
Buildkite builds:
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python, R. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java

alamb · 2022-11-28T20:47:59Z

Follow on PR: #4407

Use tournament loser tree for k-way sort-merging
5e36d5b

github-actionsBot added the core Core DataFusion crate label Nov 21, 2022

alamb requested review from alamb, tustvold and yjshen November 21, 2022 15:12

tustvold approved these changes Nov 21, 2022
View reviewed changes

viirya approved these changes Nov 21, 2022
View reviewed changes

alamb changed the title ~~Use tournament loser tree for k-way sort-merging~~Use tournament loser tree for k-way sort-merging, increase merge speed by 50%Nov 28, 2022

alamb merged commit 0d334cf into apache:masterNov 28, 2022

alamb mentioned this pull request Nov 28, 2022
Refactor loser tree code in SortPreservingMerge per PR comments #4407
Merged

Dandandan mentioned this pull request Dec 14, 2022
Add need_data_exchange in the ExecutionPlan to indicate whether a physical operator needs data exchange #4586
Merged

	letmut cmp_node = (num_streams + winner) / 2;
	// Replace overall winner by walking tree of losers
	letmut cmp_node = (num_streams + winner) / 2;

-/// Identify whether the loser tree is adjusted
+/// Identify whether the most recently yielded overall winner has been replaced
+/// within the loser tree, a value of `false` indicates that they overall winner
+/// has been yielded but the loser tree has not been updated

	match(self.is_finished(), other.is_finished()){
	// Order finished cursors last
	match(self.is_finished(), other.is_finished()){

Uh oh!

Conversation

richox commented Nov 21, 2022

Which issue does this PR close?

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

Uh oh!

alamb commented Nov 21, 2022

Uh oh!

tustvold left a comment

Choose a reason for hiding this comment

Uh oh!

tustvoldNov 21, 2022

Choose a reason for hiding this comment

Uh oh!

tustvoldNov 21, 2022

Choose a reason for hiding this comment

Uh oh!

tustvoldNov 21, 2022

Choose a reason for hiding this comment

Uh oh!

tustvoldNov 21, 2022

Choose a reason for hiding this comment

Uh oh!

tustvoldNov 21, 2022• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

tustvoldNov 21, 2022

Choose a reason for hiding this comment

Uh oh!

alambNov 28, 2022

Choose a reason for hiding this comment

Uh oh!

viiryaNov 21, 2022

Choose a reason for hiding this comment

Uh oh!

alamb commented Nov 22, 2022

Uh oh!

alamb commented Nov 26, 2022

Uh oh!

alamb commented Nov 28, 2022

Uh oh!

alamb commented Nov 28, 2022

Uh oh!

ursabot commented Nov 28, 2022

Uh oh!

alamb commented Nov 28, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

tustvoldNov 21, 2022•
edited
Loading