Messages in this thread | | | Date | Tue, 3 Nov 1998 17:17:28 +0100 | From | Pavel Machek <> | Subject | Re: [PATCH] krflushd |
| |
Hi!
> > I think that bdflush is much smarter than that: your patch makes > > kernel sync once every 30 seconds. bdflush is normally more > > intelligent: it writes smaller ammounts of data more often in order of > > not generating peaks of heavy disk i/o. > > Then, as long as bdflush is doing its smart job, there's no reason to > expect high i/o peaks.. I've been running the patch in two different > environments without any problems, so I'm happy with it.
Hugh, I thought your original goal was to kill bdflush?
Anyway, even if bdflush does it job, it looks like
Time: 0 10 20 30 40 50 seconds Write of 10M into buffers
In case of bdflush, something like Wr. 2M Wr.2M Wr.2M Wr.2M Wr.2M onto disk should happen.
In case of bdflush , something like Wr. 2M Wr.8M may happen in case of your sync awakening at t=20. Pavel
-- The best software in life is free (not shareware)! Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |