The statit output below is from a 12 disk SATA aggregate on a FAS960. The load is from a jetstress test that I let run only for a few minutes.

Disk Statistics (per second)
        ut% is the percent of time the disk was busy.
        xfers is the number of data-transfer commands issued per second.
        xfers = ureads + writes + cpreads + greads + gwrites
        chain is the average number of 4K blocks per command.
        usecs is the average disk round-trip time per 4K block.

disk    ut%  xfers  ureads--chain-usecs writes--chain-usecs cpreads-chain-usecs greads--chain-usecs gwrites-chain-usecs/aggsa/plex0/rg0:
3b.16    43  19.48    0.33   1.00 30786  17.86  30.41  1082   1.29   9.46   560   0.00   ....     .   0.00   ....     .
3b.17    43  19.43    0.33   1.00 36786  17.86  30.41  1099   1.24   8.60   485   0.00   ....     .   0.00   ....     .
2b.18    40  18.72    0.67   1.00 16893  17.43  30.88  1027   0.62  14.50   554   0.00   ....     .   0.00   ....     .
3b.19    40  18.00    0.00   ....     .  17.17  31.33  1050   0.83  10.66   566   0.00   ....     .   0.00   ....     .
2b.20    40  17.69    0.00   ....     .  17.07  31.47  1037   0.62  16.27   506   0.00   ....     .   0.00   ....     .
2b.21    42  17.67    0.00   ....     .  17.10  31.46  1097   0.57  13.54   412   0.00   ....     .   0.00   ....     .
2b.22    41  17.76    0.02   1.00 16000  17.07  31.47  1115   0.67  13.11   474   0.00   ....     .   0.00   ....     .
3b.23    40  17.91    0.00   ....     .  17.14  31.32  1045   0.76  13.53   545   0.00   ....     .   0.00   ....     .
2b.24    42  18.19    0.00   ....     .  17.29  31.11  1160   0.91   8.76  1078   0.00   ....     .   0.00   ....     .
3b.25    41  17.95    0.02   1.00  9000  17.19  31.29  1077   0.74   9.77   614   0.00   ....     .   0.00   ....     .
2b.26    43  17.88    0.00   ....     .  17.12  31.42  1167   0.76  12.72   747   0.00   ....     .   0.00   ....     .
3b.27    41  18.17    0.00   ....     .  17.19  31.25  1103   0.98  11.29   585   0.00   ....     .   0.00   ....     .

My question is, how do you interpret this data correctly? For example, was 43% of 3b.16’s time busy issuing 17.86(ignoring reads and cpreads) write commands that wrote 30.41 4KB blocks (120KB’s) and took 1082(somethings?) to return.

Is an xfer the same thing as an I/O?

What is the measurement for usecs?

Thanks for the help.

- Carl


On 3/15/06 11:24 AM, "Blake Golliher" <thelastman@gmail.com> wrote:

I ususally gauge that I can get 40 to 50 iops / sec under a 10ms
responce time.  So a little less then half of what a 10k RPM (fcal)
disk can do.

-Blake

On 3/15/06, Carl Howell <chowell@uwf.edu> wrote:
>
>
>
> Does anyone have any SATA performance data?  I know that there are many
> variables when measuring disk i/o, so any information would be appreciated.
>
>
>
> Thanks,
>
>
>
> -Carl