kerwin-zk opened a new pull request, #6380: URL: https://github.com/apache/incubator-gluten/pull/6380
## What changes were proposed in this pull request? When using ```CelebornShuffleManager```, columnar shuffle data is typically compressed on the Gluten side and sent directly to Celeborn, so compression is not needed on the Celeborn side. Therefore, ```spark.celeborn.client.shuffle.compression.codec``` is set to none. However, when row-based shuffle occurs, this configuration causes Celeborn to skip compression, leading to data bloat. ## How was this patch tested? CI -- 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]
