itskals commented on pull request #29082:
URL: https://github.com/apache/spark/pull/29082#issuecomment-675585868


   @warrenzhu25  thanks for the work... 
   I think it will add more value if we could add some quantisation fo these 
errors... So that people understand if it is really worth solving it... One 
very usefull qualification can be the "time lost" because the time spent 
earlier by a task is wasted. For example, a partition data was generated by a 
map task by running for say 1 minute...  the reducers were using this data. But 
due to some reason, the map data was lost, and the work done by the mapper 
task( 1 minute worth) had to be redone.
   If in my over all application time, if such wasted time is significant, then 
it is worth doing something to fix it.. else one could just ignore it... 
   
   @HeartSaVioR  @gengliangwang 
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to