zhouyuan commented on issue #6789: URL: https://github.com/apache/incubator-gluten/issues/6789#issuecomment-2290201093
@estoianovici The hash build allocates more memory(783G vs 1157G) although the number of input rows are the same. We have run into such issue due to huge page based allocator, but it should be fixed in the recent release. Can you please check if below config is enabled? `spark.gluten.sql.columnar.backend.velox.memoryUseHugePages` https://github.com/apache/incubator-gluten/pull/3483 thanks, -yuan -- 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]
