On 25.05.2017 0:08, Jeremy Stanley wrote: > On 2017-05-24 16:27:18 -0500 (-0500), Ben Nemec wrote: > [...] >> I spent a _lot_ of time tracking down performance issues and >> optimizing things where I could, and in the end pretty much every >> gain I made was regressed somewhere else within a few weeks, >> leaving us where we are now with jobs timing out all over the >> place. > [...] > > The uneasy truth (in many places, not just TripleO) is that if > people only try to improve speed or memory footprint or what have > you when jobs are constantly hitting those thresholds, then the > gains will usually only be very temporary. I agree it has to be an > all-the-time effort to work on minimization across these axes or > else we'll end up with near constant job failures from them. Under > development, software will (often rapidly) grow to fill the resource > constraints placed around it. There's some sort of natural law at > work there. >
Great topic, and a huge effort #deployment-time! Thank you for raising this. Few more inputs and random thoughts: [0] - Lightening undercloud MAV to save Mark Watney once again [1] - There is a strange nerdish joy with (never ending) adjusting of tests matrix and chaining builds promotion pipelines. [0] https://bugs.launchpad.net/tripleo/+bug/1693448 [1] https://bugs.launchpad.net/tripleo/+bug/1693435 -- Best regards, Bogdan Dobrelya, Irc #bogdando __________________________________________________________________________ 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