szehon-ho commented on PR #15727: URL: https://github.com/apache/iceberg/pull/15727#issuecomment-5530260376
Can we consider a configurable Spark execution mode for very large tables? The previous implementation distributed metadata grouping and joins, collecting only dangling deletes to the driver. The new implementation is more accurate, but performs full scan planning and retains applicable deletes on the driver. Do we have benchmarks or memory estimates at million-delete-file scale? -- 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]
