Did you also test indexing speed? With default G1GC settings we're seeing a slightly higher latency for queries than CMS. However, G1GC allows for much higher throughput than CMS when indexing. I haven't got the raw numbers here but it is roughly 45 minutes against 60 in favour of G1GC!
Load is obviously higher with G1GC. -----Original message----- > From:Roman Chyla <roman.ch...@gmail.com> > Sent: Wednesday 31st July 2013 18:32 > To: solr-user@lucene.apache.org > Subject: Re: Measuring SOLR performance > > I'll try to run it with the new parameters and let you know how it goes. > I've rechecked details for the G1 (default) garbage collector run and I can > confirm that 2 out of 3 runs were showing high max response times, in some > cases even 10secs, but the customized G1 never - so definitely the > parameters had effect because the max time for the customized G1 never went > higher than 1.5secs (and that happend for 2 query classes only). Both the > cms-custom and G1-custom are similar, the G1 seems to have higher values in > the max fields, but that may be random. So, yes, now I am sure what to > think of default G1 as 'bad', and that these G1 parameters, even if they > don't seem G1 specific, have real effect. > Thanks, > > roman > > > On Tue, Jul 30, 2013 at 11:01 PM, Shawn Heisey <s...@elyograg.org> wrote: > > > On 7/30/2013 6:59 PM, Roman Chyla wrote: > > > I have been wanting some tools for measuring performance of SOLR, similar > > > to Mike McCandles' lucene benchmark. > > > > > > so yet another monitor was born, is described here: > > > http://29min.wordpress.com/2013/07/31/measuring-solr-query-performance/ > > > > > > I tested it on the problem of garbage collectors (see the blogs for > > > details) and so far I can't conclude whether highly customized G1 is > > better > > > than highly customized CMS, but I think interesting details can be seen > > > there. > > > > > > Hope this helps someone, and of course, feel free to improve the tool and > > > share! > > > > I have a CMS config that's even more tuned than before, and it has made > > things MUCH better. This new config is inspired by more info that I got > > on IRC: > > > > http://wiki.apache.org/solr/ShawnHeisey#GC_Tuning > > > > The G1 customizations in your blog post don't look like they are really > > G1-specific - they may be useful with CMS as well. This statement also > > applies to some of the CMS parameters, so I would use those with G1 as > > well for any testing. > > > > UseNUMA looks interesting for machines that actually are NUMA. All the > > information that I can find says it is only for the throughput > > (parallel) collector, so it's probably not doing anything for G1. > > > > The pause parameters you've got for G1 are targets only. It will *try* > > to stick within those parameters, but if a collection requires more than > > 50 milliseconds or has to happen more often than once a second, the > > collector will ignore what you have told it. > > > > Thanks, > > Shawn > > > > >
