Why is it under-performing? this just says it executed fewer tasks,
which could be because of data locality, etc. Better is to look at
whether stages are slowed down by straggler tasks, and if so, whether
they come from one machine, and if so what may be different about that
one.
On Fri, Mar 20, 2
It seems that node is not getting allocated with enough tasks, try
increasing your level of parallelism or do a manual repartition so that
everyone gets even tasks to operate on.
Thanks
Best Regards
On Fri, Mar 20, 2015 at 8:05 PM, Yiannis Gkoufas
wrote:
> Hi all,
>
> I have 6 nodes in the clus
