zuston commented on PR #1129: URL: https://github.com/apache/incubator-uniffle/pull/1129#issuecomment-1672682714
> > Thanks for your work! This is a good progress for improving the stability of uniffle for apps. > > > During the shuffle write phase, the ShuffleServer reports faulty nodes and reallocates the ShuffleServer list; > > > > > > Not only the node failure, I hope the partition could be cached into different shuffle server if the disk is not enough or partition is huge. This has been introduce in celeborn called hard-split. > > That's another pr. We shouldn't include more too many functionities in one pr. This pr is large enough, we would better split this pr into some small prs. Agree. But I hope the design should be good enough to implement above ability. -- 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]
