Skip to content

[BLAZE-287] Integrate Blaze with Celeborn#596

Merged
richox merged 1 commit intoapache:masterfrom
RexXiong:master-integrate-celeborn-dev
Sep 27, 2024
Merged

[BLAZE-287] Integrate Blaze with Celeborn#596
richox merged 1 commit intoapache:masterfrom
RexXiong:master-integrate-celeborn-dev

Conversation

@RexXiong
Copy link
Contributor

Which issue does this PR close?

Closes #287 .

What changes are included in this PR?

Celeborn is an intermediate data service for Big Data compute engines (i.e. ETL, OLAP and Streaming engines) to boost performance, stability, and flexibility. Intermediate data typically include shuffle and spilled data. Integrate Blaze with Celeborn(0.5.1 latest stable version), will benefit Blaze + Spark in terms of shuffle performance and stability, allowing for a greater focus on optimizing computations.

Are there any user-facing changes?

No

Testing

Testing Blaze + Spark3.4+ Celeborn 0.5.1 use Some TPCDS queries with Flowing Configurations

  • spark.shuffle.manager org.apache.spark.sql.execution.blaze.shuffle.celeborn.BlazeCelebornShuffleManager
  • spark.celeborn.master.endpoints master-1-1:9097
  • spark.blaze.enable true
  • spark.sql.extensions org.apache.spark.sql.blaze.BlazeSparkSessionExtension
  • spark.memory.offHeap.enabled false

@richox richox merged commit 6918f75 into apache:master Sep 27, 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.

Supports apache-celeborn

2 participants