On Tuesday 04 January 2011 09:12:17 Duncan wrote:
> Volker Armin Hemmann posted on Tue, 04 Jan 2011 08:27:28 +0100 as
> > swap kills interactivity and is really, really bad on linux. No matter
> > how much you stripe it. Swap is a true horror. So setting swappiness to
> > 100 (which means: keep caches alive, no matter what and swap the hell
> > out of it) is a really bad idea. In my experience it is better to have a
> > very low swappiness and let the kernel get the occasional data from
> > disk, than to swap to the same disks. Strange eh?
> > But better to wait 1.5s longer for konqueror to display a directory than
> > to have a jerky mouse and input lag because of swap. Remember: the
> > kernel ALWAYS swaps out the wrong stuff.
> > I have swappiness at 60 - and back in the 4GB days at 0. Because swap
> > sucks so much.
> > You are not required to believe that. But just google.
> All I can go on is my experience, which agrees with you (and most of the
> advice on the net) when it's a single-core CPU on single-spindle storage,
> but I've found the experience rather different on multi-core machines
> driving quad-spindle striped swap on mirrored RAID, with enough memory
> swap usage is trivial in the ordinary case. Once swap usage hits half a
> gig or so, yes, it's noticed, but until then, I literally don't normally
> notice it unless/until I happen to see the usage reported on the system
> OTOH, losing portage tree cache or news (nntp) article cache and having to
> fetch the data from disk can be quite noticeable, as it DEFINITELY is the
> first time I access after a reboot (the infamous cold-cache case), the
> biggest reason I tend to leave the system running for weeks at a time,
> rebooting only to load a new test kernel or the like.
> (FWIW, I use app-admin/lib_users to track programs using stale/already-
> deleted libs after an update, and after quitting kde/X, restart daemons,
> etc, to clear the list if necessary, thus clearing that source of both
> security vulns and so-called anon-memory usage. I do use the
> portage-2.2.0_alphas with preserve-libs, but use FEATURES=-preserve-libs
> to avoid that source of bugs, so the old libs do normally get deleted.)
well, I don't waste electricity. I do have / on a ssd and /var with portage on
a raifd5. And waiting a few seconds longer for emerge -auvDtn world to give
results does not matter for me.
But a stuck keyboard because of swapping? (swap striped to three disks, 4core
And so swappiness has to stay down. Many days I don't even turn swap on.
Seeing gcc oom is much less annoying than swap.
It got better over the years but it is still far from being acceptable.