High CPU system usage

[Glenn Steen]

    [ The following text is in the "ISO-8859-1" character set. ]
    [ Your display is set for the "US-ASCII" character set.  ]
    [ Some characters may be displayed incorrectly. ]

On 24/08/05, Ugo Bellavance <ugob at camo-route.com> wrote:
> Glenn Steen wrote:
> > On 23/08/05, Ugo Bellavance <ugob at camo-route.com> wrote:
> > (snip)
> >
> >>10:40:00 AM         0     77.90      0.00     15.45      0.58      6.07
> >>10:40:00 AM         1     77.85      0.00     15.38      0.59      6.18
> >>10:49:59 AM       all     65.56      0.00     17.65      3.93     12.87
> >>10:49:59 AM         0     64.74      0.00     17.76      3.89     13.61
> >>10:49:59 AM         1     66.37      0.00     17.54      3.96     12.13
> >>11:00:01 AM       all     64.52      0.00     22.36      6.82      6.30
> >>11:00:01 AM         0     64.78      0.00     22.18      6.75      6.28
> >>11:00:01 AM         1     64.26      0.00     22.55      6.89      6.31
> >>11:10:01 AM       all     64.00      0.00     21.29      6.63      8.08
> >>11:10:01 AM         0     63.83      0.00     21.29      6.66      8.23
> >>11:10:01 AM         1     64.17      0.00     21.30      6.60      7.93
> >>11:20:00 AM       all     69.71      0.00     21.51      3.57      5.21
> >>11:20:00 AM         0     69.45      0.00     21.55      3.62      5.39
> >>11:20:00 AM         1     69.98      0.00     21.47      3.52      5.03
> >
> > (snip)
> > Your %iowait seem to be a bit high perhaps (could be normal:-) .... If
> > you look at the "-d" section for your block device, does it seem to be
> > close to a .... "measurable max"?
> >
> you mean, this part?
> 
> 07:40:00          tps      rtps      wtps   bread/s   bwrtn/s
> 09:20:00        55.68      2.54     53.14     27.67   2181.37
> 09:30:00        37.81      2.17     35.64     23.16   1569.62
> 09:40:01        45.35      1.95     43.41     21.06   1758.61
> 09:50:01        73.55      2.71     70.85     32.03   2121.82
> 10:00:00       100.95      4.75     96.20     51.44   2297.69
> 10:10:01        58.90      2.01     56.90     22.15   2480.38
> 10:20:01        40.88      1.32     39.56     14.40   1889.31
> 10:30:01        36.79      1.24     35.54     13.49   1597.77
> 10:40:00        46.61      1.51     45.10     16.32   2276.00
> 10:49:59       125.51      4.60    120.90     52.40   2598.60
> 11:00:01       232.63      8.64    223.98     97.07   2778.52
> 11:10:01       222.20      7.29    214.91     82.03   2659.14
> 11:20:00       161.06      5.30    155.76     62.35   2268.54
> 11:30:00        93.75      2.35     91.39     24.82   1940.26

Not really, although this should do... I was thinking of

00:01:01          DEV       tps  rd_sec/s  wr_sec/s
00:11:01       dev3-0      3.24      1.93    112.65
....
which perhaps works better in 1.6 anyway, so...
You seem to have a "write peak" at ~ 1.36 MiB/s, which shouldn't be
partcularily close to your theoretical max, ergo not likely to be
close to your real max (whatever that is:-).

> 
> I can't really think of a way to determine this 'measurable max'.
> HDparm?  Analysis of a sar report while testing hard drive?

More or less so, yes. Also try using iostat.
Perhaps add in Bonnie there too
(http://www.garloff.de/kurt/linux/bonnie/)... and perhaps eye through
http://www.netlib.org/performance/papers/bonnie/ ... disk and fs
performance measuring isn't easy;-).

<thinking_out_load>
But... I don't *think* you're I/O-starved. That top someone sugegsted
should probably show what's up (or at least "who":).