apache/beam

Multiple batch joins on the same key results in multiple shuffles

Open

#19,097 opened on Jun 3, 2022

View on GitHub
 (0 comments) (0 reactions) (0 assignees)Java (4,097 forks)batch import
P3dslgood first issueimprovementsql

Repository metrics

Stars
 (7,313 stars)
PR merge metrics
 (Avg merge 6d) (243 merged PRs in 30d)

Description

The query like this:


SELECT a.*, b.*, c.* FROM a JOIN b ON a.some_id = b.some_id JOIN c ON a.some_id = c.some_id;

results in two shuffles. Can probably be optimized.

Relevant code:

Imported from Jira BEAM-5049. Original Jira may contain additional context. Reported by: kedin.

Contributor guide

Research direction
Examine the CoGBK join implementation and BeamJoinRel to understand why multiple shuffles occur for joins on the same key. Look for opportunities to combine shuffles by reusing the partitioned data from the first join.
Tech stack
java
Domain
backenddata
Issue type
Performance
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
JavaApache BeamSQL
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
40