Wish I had an answer for you.. This is relevant to our interests as well as I planned to run spark jobs on our slave cluster. If we don't hear conclusively on the list I will do some benchmarking in a dev environment and attempt to reproduce
Best, Iain Sent from my iPhone > On Jul 15, 2016, at 5:01 PM, Saad Mufti <[email protected]> wrote: > > Hi, > > Don't have anything conclusive but I have seen some correlation where in > very high write rate situation, the write rate can increase when major > compaction or some other high CPU/network activity (for example we run some > Spark jobs on our replica HBase cluster) stops happening on the replica > cluster. Basically looks like the rate is lower when the replica cluster is > busy with some other activity. > > I wanted to know if this totally anecdotal evidence is something I should > look into or whether it is something well known in the community. > > Thanks in advance for any pointer or advice for someone relatively new to > HBase. > > ---- > Saad
