I added the patch to the branch as you suggested Andrew (Tests passed for me locally). St.Ack
On Fri, Jun 10, 2011 at 3:18 AM, Andrew Purtell <[email protected]> wrote: > Stack, > > Aside from the other ideas you mention, this could also be HBASE-3855 if a > lot of values are up in memstore? > > By the way, I patched our 0.90-ish in house HBase with 3855 with no untoward > affects and noticeable improvement under profiling. Why not commit it to 0.90 > as well? > > Best regards, > > - Andy > > Problems worthy of attack prove their worth by hitting back. - Piet Hein (via > Tom White) > > > --- On Wed, 6/8/11, Stack <[email protected]> wrote: > >> From: Stack <[email protected]> >> Subject: Re: increasing hbase get latencies >> To: [email protected] >> Date: Wednesday, June 8, 2011, 11:31 PM >> On Wed, Jun 8, 2011 at 4:39 PM, >> Abhijit Pol <[email protected]> >> wrote: >> > Recently we observed that our "get" latencies keep increasing >> > over the period (and eventually flatten out at higher value) >> > and if we restart hbase server, latencies go back to good >> > state (low values) and start increasing again. > >
