Skip to content
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

[CORE] ColumnarBroadcastExchangeExec should set/cancel with job tag for Spark3.5 #4882

Merged
merged 1 commit into from
Mar 8, 2024

Conversation

ulysses-you
Copy link
Contributor

What changes were proposed in this pull request?

Spark3.5 uses job tag to cancel broadcast exchange job. This pr adds two shim methods to be compatible with that:

setJobDescriptionOrTagForBroadcastExchange
cancelJobGroupForBroadcastExchange

See apache/spark#41440

How was this patch tested?

Pass CI

Copy link

github-actions bot commented Mar 7, 2024

Thanks for opening a pull request!

Could you open an issue for this pull request on Github Issues?

https://github.com/oap-project/gluten/issues

Then could you also rename commit message and pull request title in the following format?

[GLUTEN-${ISSUES_ID}][COMPONENT]feat/fix: ${detailed message}

See also:

Copy link

github-actions bot commented Mar 7, 2024

Run Gluten Clickhouse CI

runId.toString,
s"broadcast exchange (runId $runId)",
interruptOnCancel = true)
SparkShimLoader.getSparkShims.setJobDescriptionOrTagForBroadcastExchange(sparkContext, this)
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This change is required, otherwise AQE can not cancel broadcast job. BroadcastQueryStageExec use cancelJobsWithTag to cancel broadcast.

Copy link

github-actions bot commented Mar 8, 2024

Run Gluten Clickhouse CI

@ulysses-you
Copy link
Contributor Author

cc @PHILO-HE @JkSelf thank you

Copy link
Contributor

@JkSelf JkSelf left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. Thanks.

@JkSelf JkSelf merged commit 9adc18d into apache:main Mar 8, 2024
17 checks passed
@ulysses-you ulysses-you deleted the broadcast branch March 8, 2024 12:09
@GlutenPerfBot
Copy link
Contributor

===== Performance report for TPCH SF2000 with Velox backend, for reference only ====

query log/native_4882_time.csv log/native_master_03_06_2024_bddc3fd79_time.csv difference percentage
q1 35.68 38.77 3.097 108.68%
q2 25.57 24.36 -1.216 95.24%
q3 37.07 39.60 2.532 106.83%
q4 39.00 37.58 -1.417 96.37%
q5 71.05 69.92 -1.128 98.41%
q6 5.88 8.34 2.451 141.65%
q7 85.66 84.40 -1.265 98.52%
q8 84.27 86.73 2.460 102.92%
q9 120.41 119.49 -0.921 99.24%
q10 45.40 43.18 -2.217 95.12%
q11 20.35 20.86 0.513 102.52%
q12 29.00 28.05 -0.942 96.75%
q13 48.60 44.55 -4.056 91.66%
q14 17.12 17.03 -0.087 99.49%
q15 32.37 28.15 -4.225 86.95%
q16 14.02 14.07 0.058 100.41%
q17 101.90 101.46 -0.436 99.57%
q18 141.47 145.26 3.792 102.68%
q19 13.71 13.99 0.276 102.01%
q20 26.75 28.21 1.460 105.46%
q21 229.84 224.18 -5.660 97.54%
q22 13.86 14.99 1.130 108.16%
total 1238.98 1233.17 -5.803 99.53%

taiyang-li pushed a commit to bigo-sg/gluten that referenced this pull request Mar 25, 2024
taiyang-li pushed a commit to bigo-sg/gluten that referenced this pull request Oct 8, 2024
taiyang-li pushed a commit to bigo-sg/gluten that referenced this pull request Oct 9, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants