petern48 commented on PR #171: URL: https://github.com/apache/sedona-db/pull/171#issuecomment-3466172666
FYI, our performance compared to duckdb is not much farther ahead as shown in the original benchmark in the PR description. Here's the updated benchmark. <img width="814" height="218" alt="image" src="https://github.com/user-attachments/assets/8e93e2af-45e0-4ab3-8658-129acfa19d76"; /> Main differences are: 1) This updated benchmark uses the single-threaded Sedona. I'm guessing this is the major cause of the perf drop. 2) We've implemented the full WkbHeader since the original benchmark. We do more than the original implementations did, so it should be slower. Mainly, we support EWKB and SRID checks, and we also grab `first_xy`. `first_xy` shouldn't be much more effort since we already were getting to first_geom for the `first_geom_dimensions` field. I've addressed all main slowdowns I was aware of (e.g. creating nested buffers), but I wonder if there are more. -- 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]
