andygrove opened a new pull request, #2161:
URL: https://github.com/apache/datafusion-ballista/pull/2161
## Summary
Refresh `docs/source/contributors-guide/benchmarking.md` with a new TPC-H
**SF1000** result set captured on a 32-executor Kubernetes cluster with
Parquet data on S3.
- Pins the exact commit under test: `696ca29b` (`main`, 2026-07-23,
Ballista `54.0.0`, DataFusion `54.1.0`).
- Documents the environment (cluster shape, executor sizing) and the full
Ballista + Spark configurations used.
- Publishes Q1–Q19 mean times over 2 iterations; marks Q20–Q22 as `FAIL`
on this configuration.
- Adds a vanilla **Spark 3.4** baseline column measured on the same
cluster shape, alongside the Ballista column.
- Trims the surrounding narrative from the previous homelab-era version of
the page so the document mainly states current performance.
## Notes
- Ballista's `Q18` now completes at this commit and sizing (399.72 s).
- Q20–Q22 did not produce results in this run; recorded as `FAIL`
pending root-cause investigation.
- Spark row uses the harness's mean-of-3 (first iteration dropped);
Ballista row is mean-of-2.
## Test plan
- [x] Doc renders as GitHub markdown (tables + code blocks).
- [ ] Reviewer sanity-check: numbers match the JSON captured for this run.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
