1 |
On Wednesday 29 November 2006 22:17, "Vladimir G. Ivanovic" |
2 |
<vgivanovic@×××××××.net> wrote about 'Re: [gentoo-user] [OT] HDD I/O and |
3 |
UI responsiveness': |
4 |
> Boyd Stephen Smith Jr. wrote: |
5 |
> > On Tuesday 28 November 2006 17:45, Dale <dalek@××××××××××.net> wrote |
6 |
> > about 'Re: [gentoo-user] [OT] HDD I/O and UI responsiveness': |
7 |
> >> I read somewhere that they are trying to 'nice' the drive usage like |
8 |
> >> they do the CPU. That may help if you can find it and enable it. I |
9 |
> >> think it is in some of the very new kernels if I read it correctly. |
10 |
> >> Sorry, I didn't bookmark it. < slaps hand > |
11 |
> > |
12 |
> > There's the ionice utility provided by... |
13 |
> > <Running equery> |
14 |
> > ...sys-process/schedutils. |
15 |
> |
16 |
> Isn't this just masking the problem? |
17 |
|
18 |
Yes and no. |
19 |
|
20 |
> I don't see the CPU pegged, so why should other applications be |
21 |
> unresponsive? |
22 |
|
23 |
Because they aren't waiting to be scheduled on the CPU. They are waiting |
24 |
to be scheduled for I/O against a certain device. ionice will cause a |
25 |
process to be more likely to have interruptable I/O operations |
26 |
interrupted, allowing other non-ioniced processes to access the disk(s) |
27 |
with less latency. |
28 |
|
29 |
Dumping the contents of /dev/null to disk will slow other processes down |
30 |
even though your CPU will be used very little, because all your I/O is |
31 |
consumed. |
32 |
|
33 |
It is "masking" the problem that you are running out of I/O |
34 |
bandwidth / "latency" / "cycles". Of course, that's not actually that |
35 |
rare; I have a 6 disk RAID-5 array, and I can still make playing video |
36 |
from disk stutter, by performing I/O intensive tasks. |
37 |
|
38 |
-- |
39 |
"If there's one thing we've established over the years, |
40 |
it's that the vast majority of our users don't have the slightest |
41 |
clue what's best for them in terms of package stability." |
42 |
-- Gentoo Developer Ciaran McCreesh |