[ 
https://issues.apache.org/jira/browse/IMPALA-15097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18088529#comment-18088529
 ] 

Aleksandr Efimov commented on IMPALA-15097:
-------------------------------------------

I’ve been working with Impala profile parsing/diagnostic tooling recently, and 
runtime filters are an area where we’ve hit a few practical edge cases. I’d 
like to make the first patch a bit more useful than only drawing dashed lines.

Proposed initial scope:
- mark runtime-filter consumer edges as dashed when the profile shows that the 
filter processed input but rejected nothing;
- show compact effectiveness counters on the graph when available, e.g. 
rows/files/row-groups rejected out of processed/total;
- read the counters from the scan-node runtime filter sections, while keeping 
the graph unchanged when the profile does not contain enough information;
- make sure the parser handles the current runtime filter table format, 
including the effective-targets column;
- add a focused regression test with a small profile snippet, and verify the 
output on the attached profile.

I’ll keep broader profile-graph improvements as follow-ups so this issue stays 
focused on runtime filter effectiveness.

> Show runtime filter effectiveness in plan-graph.py
> --------------------------------------------------
>
>                 Key: IMPALA-15097
>                 URL: https://issues.apache.org/jira/browse/IMPALA-15097
>             Project: IMPALA
>          Issue Type: Improvement
>            Reporter: Quanlong Huang
>            Assignee: Aleksandr Efimov
>            Priority: Major
>         Attachments: query-plan1.svg, rf_profile.txt
>
>
> When verbosity is larger than 0, plan-graph.py generates nodes and lines for 
> runtime filters. It'd be helpful to show ineffective runtime filters, e.g., 
> by using dash lines.
> It'd be more helpful if we should how many files/rows, etc. are skipped by 
> the filter.
> Uploaded an example profile and the generated svg file. The command is
> {code:java}
> bin/diagnostics/experimental/plan-graph.py -v1 rf_profile.txt | dot -T svg -o 
> query-plan1.svg {code}



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to