Re: Integrating ML/DL frameworks with Spark

Hi all, Thanks for your feedback! I uploaded a SPIP doc for the barrier scheduling feature at https://issues.apache.org/jira/browse/SPARK-24374. Please take a look and leave your comments there. I had some offline discussion with +Xingbo Jiang to help me design the

Re: [VOTE] Spark 2.3.1 (RC2)

Sure. Also, I'd appreciate if these bugs were properly triaged and targeted, so that we could avoid creating RCs when we know there are blocking bugs that will prevent the RC vote from succeeding. On Wed, May 23, 2018 at 9:02 AM, Xiao Li wrote: > -1 > > Yeah, we should fix

Re: [VOTE] Spark 2.3.1 (RC2)

-1 Yeah, we should fix it in Spark 2.3.1. https://issues.apache.org/jira/browse/SPARK-24257 is a correctness bug. The PR can be merged soon. Thus, let us have another RC? Thanks, Xiao 2018-05-23 8:04 GMT-07:00 chenliang613 : > Hi > > Agree with Wenchen, it is better

Re: [VOTE] Spark 2.3.1 (RC2)

Hi Agree with Wenchen, it is better to fix this issue. Regards Liang cloud0fan wrote > We found a critical bug in tungsten that can lead to silent data > corruption: https://github.com/apache/spark/pull/21311 > > This is a long-standing bug that starts with Spark 2.0(not a regression), > but

Re: [VOTE] Spark 2.3.1 (RC2)

We found a critical bug in tungsten that can lead to silent data corruption: https://github.com/apache/spark/pull/21311 This is a long-standing bug that starts with Spark 2.0(not a regression), but since we are going to release 2.3.1, I think it's a good chance to include this fix. We will also

Re: [VOTE] Spark 2.3.1 (RC2)

+1 Same result for me as with RC1. On Tue, May 22, 2018 at 2:45 PM Marcelo Vanzin wrote: > Please vote on releasing the following candidate as Apache Spark version > 2.3.1. > > The vote is open until Friday, May 25, at 20:00 UTC and passes if > at least 3 +1 PMC votes are