To your other question we're still trying to come up with numbers, the system
has many things going on and showing something 100% attributable to this is
hard - I'm trying to see if we can get a more exact A/B test.
madvise() doesn't help ensure there's less dirty data in the system, which is
different from what you are describing. If the write-queue becomes too long,
then anything else that needs to write now has to wait... which doesn't work
well with a database load. Now mysql sync's and has to wait for a half-gig of
data to be flushed. That's a problem.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/187#issuecomment-292278640
_______________________________________________
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint