JoshRosen opened a new pull request #27089: [SPARK-30414][SQL][WIP] ParquetRowConverter optimizations: arrays, maps, and constant factors URL: https://github.com/apache/spark/pull/27089 <!-- Thanks for sending a pull request! Here are some tips for you: 1. If this is your first time, please read our contributor guidelines: https://spark.apache.org/contributing.html 2. Ensure you have added or run the appropriate tests for your PR: https://spark.apache.org/developer-tools.html 3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][SPARK-XXXX] Your PR title ...'. 4. Be sure to keep the PR description updated to reflect all changes. 5. Please write your PR title to summarize what this PR proposes. 6. If possible, provide a concise example to reproduce the issue for a faster review. --> ### What changes were proposed in this pull request? <!-- Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue. If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below. 1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers. 2. If you fix some SQL features, you can provide some references of other DBMSes. 3. If there is design documentation, please add the link. 4. If there is a discussion in the mailing list, please add the link. --> This PR implements multiple performance optimizations for `ParquetRowConverter`, achieving some modest constant-factor wins for all fields and larger wins for map and array fields: - Add `private[this]` to several `val`s (90cebf080a5d3857ea8cf2a89e8e060b8b5a2fbf) - Keep a `fieldUpdaters` array, saving two`.updater()` calls per field (7318785d350cc924198d7514e40973fd76d54ad5): I suspect that these are often megamorphic calls, so cutting these out seems like it could be a relatively large performance win. - Only call `currentRow.numFields` once per `start()` call (e05de150813b639929c18af1df09ec718d2d16fc): previously we'd call it once per field and this had a significant enough cost that it was visible during profiling. - Re-use buffers in array and map converters (c7d1534685fbad5d2280b082f37bed6d75848e76): previously we would create a brand-new Scala `ArrayBuffer` for each field read, but this isn't actually necessary because the data is already copied into a fresh array when `end()` constructs a `GenericArrayData`. Here, I've replaced this with re-used Java `ArrayList`s. ### Why are the changes needed? <!-- Please clarify why the changes are needed. For instance, 1. If you propose a new API, clarify the use case for a new API. 2. If you fix a bug, you can clarify why it is a bug. --> To improve Parquet read performance; this is complementary to #26993's (orthogonal) improvements for nested struct read performance. ### Does this PR introduce any user-facing change? <!-- If yes, please clarify the previous behavior and the change this PR proposes - provide the console output, description and/or an example to show the behavior difference if possible. If no, write 'No'. --> No. ### How was this patch tested? <!-- If tests were added, say they were added here. Please make sure to add some test cases that check the changes thoroughly including negative and positive cases if possible. If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future. If tests were not added, please describe why they were not added and/or why it was difficult to add. --> I'm marking this as `[WIP]` pending more benchmarking. Early benchmarks with both synthetic and realistic schemas (similar to the ones in #26993) suggest moderate gains (10%+ in certain cases).
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
