apache / apache/gluten

TPC-DS part of SQL's broadcasthashjoin metric is inconsistent

Open
#9,423 4 comments 0 reactions 0 assignees View on GitHub
bug triage
Dominant language
Scala
Stars
1.6k
Forks
657
Avg merge
2d 21h
Merged PRs (30d)
85

Description

### Backend

VL (Velox)

### Bug description

The metric of BroadcastHashJoinExecTransformer "number of hash build input rows" different from the InputIteratorTransformer "number of output rows".

The web ui is tpc-ds sql12:
![Image](https://github.com/user-attachments/assets/c5ff65fb-5133-41c0-8de3-8686eceaa7da)

### Gluten version

Gluten-1.3

### Spark version

Spark-3.3.x

### Spark configurations

_No response_

### System information

_No response_

### Relevant logs

```bash

```

Contributor guide

Open the contributing guide

Research direction

Start with the TPC-DS SQL12 web UI evidence and compare the metric reported by BroadcastHashJoinExecTransformer, "number of hash build input rows," with InputIteratorTransformer's "number of output rows." The issue is complete when the cause of the discrepancy is identified and the two metrics are consistent or their intended difference is documented.

Written by the indexing model from the issue text.

Assessment

Tech stack
sql
Domain
backend, databases
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.