It is based on the number of live regions. Jerry
On Fri, Oct 21, 2016 at 7:50 AM, Vadim Vararu <vadim.var...@adswizz.com> wrote: > Hi guys, > > I'm trying to run the importTSV job and to write the result into a remote > HDFS. Isn't it supposed to write data concurrently? Asking cause i get the > same time with 2 and 4 nodes and i can see that there is only 1 reduce > running. > Where is the bottleneck? > > Thanks, Vadim. >