You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kylin.apache.org by ć‘šćƒæ˜Š <qh...@apache.org> on 2015/08/19 10:55:23 UTC

benchmark setup for spark & mr

Hi,
    I am going to benchmark cubing on spark and cubing on mr.
    But first I have to setup separate cluster for both of them.
    Are there any suggestions how the cluster should be setup, so that we
can get a fair competition?