Subscribe

Is there any command to get the utilization of disk which is part of aggregate?

Don't want aggregate and volume usage. But looking for utilization of all disks separately in that aggregate.

Re: Is there any command to get the utilization of disk which is part of aggregate?

You can use “stats” command to get large number of counters including individual disks.

Re: Is there any command to get the utilization of disk which is part of aggregate?

Hi Rajiv,

If you are looking for indivisual disk space utilization in an aggregate, thats not possible.

you can report on aggregatge usage only (capacity)

If you are looking on IOPs per disk that can be checked via

- perfstat

- statit

- sysstat will show overall disk utilization

Nitish 

Re: Is there any command to get the utilization of disk which is part of aggregate?

If you are looking for indivisual disk space utilization in an aggregate, thats not possible.

Just a thought - if there is a concern about non-balanced disk utilisation, reallocate command may help: both with measuring & addressing the issue.

Regards,
Radek

Re: Is there any command to get the utilization of disk which is part of aggregate?

I stumbled across this thread. You can get the disk utilization for all the disks in a given aggregate.  Use Performance Advisor, which is part of OnCommand 5.1 Unified Manager.  This software is free to all NetApp customers (new and existing).  You can find it on support.netapp.com under the software download section.

In Performance Advisor, you can navigate to the desired controller and view the aggregates that belong to it.  All of the disk drives that make-up that aggregate are listed below it.  You can either view each individual disk drive's utilization or you can view the top-5 busiest dirves.  You can also make a custom view to show the utilization of all disk drives for the aggregate on a single chart.  Fairly easy to do.

Re: Is there any command to get the utilization of disk which is part of aggregate?

If you're after IOPS on each disk then the easiest way to see them all at a glance is with 'statit'.

Run 'statit -b', wait a period of time for it to gather data (30-60 seconds), then run statit -e.

Make sure you have a wide CLI session with sufficient buffering/logging. Look for the section which looks like this:

                       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

/aggr0/plex0/rg0:

0a.27              1   1.26    0.32   1.00 17000   0.95   1.00     0   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.26              1   1.26    0.32   1.00 17000   0.95   1.00     0   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.28              2   2.52    2.52   1.00 24500   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.29              2   2.21    2.21   1.00 18000   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.22              1   1.89    1.89   1.00 14167   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.21              1   0.95    0.95   1.00 17000   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.20              1   1.26    1.26   1.00 18000   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.19              2   1.26    1.26   1.00 49000   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.25              2   0.63    0.63   1.00 92500   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.17              1   1.89    1.89   1.00 14667   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.24              2   2.52    2.52   1.00 25375   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

0a.16              3   1.58    1.58   1.00 58600   0.00   ....     .   0.00   ....     .   0.00   ....     .   0.00   ....     .

Aggregate statistics:

Minimum            1   0.63    0.32                0.00                0.00                0.00                0.00

Mean               2   1.58    1.26                0.00                0.00                0.00                0.00

Maximum            3   2.52    2.52                0.95                0.00                0.00                0.00

This is a very quiet aggregate with only 1 raidgroup, but you will be able to see all of your disks and compare their IOPS.

This is a much easier way to see the performance of all disks at a single point in time rather than using Performance Advisor (where you need to view each disk separately)... Performance Advisor is better if you want to see individual disk performance over time.

Cheers,

John