zhengchenyu opened a new issue, #1249: URL: https://github.com/apache/incubator-uniffle/issues/1249
### Code of Conduct - [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct) ### Search before asking - [X] I have searched in the [issues](https://github.com/apache/incubator-uniffle/issues?q=is%3Aissue) and found no similar issues. ### Describe the feature https://github.com/apache/incubator-uniffle/commit/8683254720974c7254f9c5c2fd902743ae0ab273 disable write data to MR's OnDiskMapOutput. So it means we can not support the large shuffle data which is over mapreduce.reduce.shuffle.memory.limit.percent * memoryLimit. In this issue, will enable this function. Here's the key point: RssFetcher will fetch compressed data(compressed by uiffle), then decompressed to the parameter 'buffer'. But when MergeManagerImpl merge the temporary shuffle data, will use hadoop compression code. ### Motivation _No response_ ### Describe the solution _No response_ ### Additional context _No response_ ### Are you willing to submit PR? - [X] Yes I am willing to submit a PR! -- 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]
