Re: Reducing HDD writing affect on whole system.
Thank You for Your time and answer, Henrique:
>Well, *any* issue with the disk subsystem will cause such problems to
>get several orders of magnitude worse, so yes, one must *first* make
>sure the disks are operating at the expected speed.
How I can be sure the HDD is "operating at the expected speed"?
>After you are sure the hardware is operating at its proper speed, you
>can start tunning the kernel and IO tasks. There is indeed such a
>thing as an IO priority per process, but due to CFQ/writeback issues,
>tuning that might not be enough. You also have to tune the kernel to
>not leave too many outstanding pages. And some filesystems are better
>than others for certain workloads.
Let me make it clearer a bit *my* situation.
For now I'm speaking about simple desktop environment - nothing special
w/ the HDD/FS - just SATA disk w/ EXT4 on it. I experience freezes on
any applications while, say, I copy a a big file - say just a DVD.iso -
I do not think it is normal absolutely. And as You pointed out that the
discs can work not at their optimal speed (hardware issue) then I would
like to find out that. - I have read that the devices do tune
themselves pretty well for the optimal performance, though. So, how I
check that?
>ionice from util-linux can set the process IO priority. There are
>other utilites that can also do that.
Is it "nice" :) if I put the following
ionice -c 2 -n 7 mc
to /etc/rc.local
? Or should put it in some other place - the idea is to run all the mc
processes w/ lower IO priorities.
>I couldn't readly find any up-to-date (well, up-to-2.6.32) guide on how
>to tune the VM subsystem (which controls the writeback) to refer you
>to, please look for documentation on how to mess with
>the /proc/sys/vm/* tunables.
OK.
>And there is always the possibility that you have to actually tune the
>applications. Even if the filesystem, VM and IO scheduler were perfect
>(and they're so far from being perfect it is not funny) you'd still
>have only so much IO bandwidth. Also, data access patterns can matter
>a lot (even on SSDs, when there are write operations).
In other words we do not have much to do, right?
Reply to: