voonhous opened a new pull request, #20647: URL: https://github.com/apache/hudi/pull/20647
### Describe the issue this Pull Request addresses Closes #20626. Part of #20625. MOR snapshot read work in the Trino connector had no way to be measured without a cluster and an object store. ### Summary and Changelog Adds two on-demand benchmarks to `hudi-trino` test scope: - `HudiMergeBufferBenchmark`: pushes one file group's log records through the spillable merge map as sized by the Trino reader context; reports estimate per record, spilled count, put and remove cost. - `HudiMorReadBenchmark`: runs `_ro` and `_rt` queries on a generated TPC-DS-shaped MOR table behind a file system wrapper that injects object-store latency and counts requests per file type. Both assert their own results against the generator's expected values. The classes are named `*Benchmark`, so surefire and CI never run them. `hudi-trino/README.md` gains a Benchmarks section with the two run commands and the `bench.*` knobs. ### Impact Test scope only. No production code changes. ### Risk Level none ### Documentation Update README section in `hudi-trino`. ### Contributor's checklist - [x] Read through [contributor's guide](https://hudi.apache.org/contribute/how-to-contribute) - [x] Enough context is provided in the sections above - [x] Adequate tests were added if applicable -- 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]
