hi folks, i didn't get as far as i'd hope so i've decided to release what i have currently and create another deck for more 'future enhancements' benchmarking.
the following deck aims to show how performance changes as you scale out Ceph and some configuration options that can help stabilise Gnocchi+Ceph performance: http://www.slideshare.net/GordonChung/gnocchi-profiling-v2. this is by no means a large scale architecture -- the tests are run against tens of thousands metrics currently -- but it's a start until we get more data. i'm hoping to test against a larger dataset going forward. also, will be testing some enhancements we've been discussing for Gnocchi 3.x hope it helps. cheers, On 25/06/2016 8:50 AM, Curtis wrote: > On Fri, Jun 24, 2016 at 2:09 PM, gordon chung <g...@live.ca> wrote: >> hi, >> >> i realised i didn't post this beyond IRC, so here are some initial >> numbers for some performance/benchmarking i did on Gnocchi. >> >> http://www.slideshare.net/GordonChung/gnocchi-profiling-21x >> >> as a headsup, the data above is using Ceph and with pretty much a >> default configuration. i'm currently doing more tests to see how it >> works if you actually start turning some knobs on Ceph (spoiler: it gets >> better). > > Thanks Gordon. I will definitely take a look through your slides. > Looking forward to what test results you get when you start turning > some knobs. > > Thanks, > Curtis. > >> >> cheers, >> >> -- >> gord >> _______________________________________________ >> OpenStack-operators mailing list >> openstack-operat...@lists.openstack.org >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators > > > -- gord __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
