How do I test NFM-P disk performance?

Purpose

Perform this procedure to check the disk performance on an NFM-P component station.

The disk performance of an NFM-P component affects overall system performance, and must meet or exceed the minimum specifications in the response to the NFM-P Platform Sizing Request for your system. See the NSP NFM-P Release Description for information about submitting a Platform Sizing Request.

Also, before you add capacity to a disk or partition on an NFM-P component, for example, using LVM, you must ensure that the disk throughput and latency values remain within tolerance, which is defined as being within 10% of the current values.

CAUTION 

CAUTION

Service Disruption

Checking NFM-P disk performance requires a shutdown of one or more NFM-P components, which is service-affecting.

Perform the procedure only during a scheduled maintenance period.

Steps
 

Perform one of the following, depending on the type of component for which you need to check performance:

  1. Shut down a main server.

    1. Log in to the main server station as the nsp user.

    2. Open a console window.

    3. Navigate to the /opt/nsp/nfmp/server/nms/bin directory.

    4. Enter the following:

      bash$ ./nmsserver.bash stop ↵

    5. Enter the following to display the server status:

      bash$ ./nmsserver.bash appserver_status ↵

      The server status is displayed; the server is fully stopped if the status is the following:

      Application Server is stopped

      If the server is not fully stopped, wait five minutes and then repeat this step. Do not perform the next step until the server is fully stopped.

  2. Shut down an auxiliary server.

    1. Log in to the auxiliary server station as the nsp user.

    2. Open a console window.

    3. Navigate to the /opt/nsp/nfmp/auxserver/nms/bin directory.

    4. Enter the following:

      bash$ ./auxnmsserver.bash auxstop ↵

    5. Enter the following to display the auxiliary server status:

      bash$ ./auxnmsserver.bash auxappserver_status ↵

      The command returns a status message.

    6. The server is fully stopped when the following is displayed:

      Auxiliary Server is stopped

      If the server is not fully stopped, wait five minutes and then repeat this step. Do not perform the next step until the server is stopped.

  3. Shut down a main database.

    1. Log in to the main database station as the root user.

    2. Open a console window.

    3. Enter the following to stop the Oracle proxy:

      systemctl stop nfmp-oracle-proxy.service ↵

    4. Enter the following to stop the main database:

      systemctl stop nfmp-main-db.service ↵

  4. Shut down an auxiliary database; perform How do I stop an auxiliary database cluster?.


If you are performing the test on a main or auxiliary server station, enter the following to switch to the root user:

bash$ su - ↵


Perform one of the following.

  1. On a main server station, enter the following:

    /opt/nsp/nfmp/server/nms/bin/unsupported/IOTest/NSP_IOTest.pl -d target

    where target is the disk partition to test

  2. On an auxiliary server station, enter the following:

    /opt/nsp/nfmp/auxserver/nms/bin/unsupported/IOTest/NSP_IOTest.pl -d target

    where target is the disk partition to test

  3. On a main database station, enter the following:

    /opt/nsp/nfmp/db/install/tools/unsupported/IOTest/NSP_IOTest.pl -d target

    where target is the disk partition to test


Record the utility output.


If you are performing the test as a pre-upgrade task specified in the NSP Installation and Upgrade Guide, or as a general performance check, perform the following steps.

  1. Compare the following recorded values with the values specified for your system:

    • main server, main database, or auxiliary server—Read, Write, and Latency

    • NSP auxiliary database—Read, Write, Rewrite, and %IO Wait

  2. If the values do not meet the minimum specifications, contact the NFM-P Platform Team through your account representative.

  3. Go to Step 7.


If you are adding capacity to a disk or partition, perform the following steps.

  1. Add the required capacity to the disk or partition.

  2. Repeat Step 3 and Step 4 as required.

  3. Compare the following values from before and after the capacity increase:

    • main server, main database, or auxiliary server—Read, Write, and Latency

    • auxiliary database—Read, Write, Rewrite, and %IO Wait

  4. If the values differ by more than 10%, contact the NFM-P Platform Team through your account representative.


Close the console windows, as required.

End of steps