Calculating throughput for IBM Storage Protect database disks

You can use benchmarking tools to measure throughput capability of the IBM Storage Protect database disks during full database backup operations.

1. Use the tsmdiskperf.pl Perl script as a benchmarking tool to determine the input/output operations per second (IOPS) capability of the database disks:
   1. Benchmark the directory paths that are used for the read-only workload of database sequential disks with a 256 KB block size.
   2. To run the script, issue the following command:

      perl tsmdiskperf.pl workload=stgpool mode=readonly fslist=directory_list

      where directory_list is a comma-separated list of database directories.
      Tip: Specify the workload=stgpool parameter to ensure that sequential reads occur during database backup operations.
   3. Ensure that the read-only data ingestion rate that is obtainable for these database paths meets the speed requirements for full database backups to finish within the scheduled timeframe.
   For benchmarking tools and sample benchmarking tests, see the IBM Storage Protect Blueprint. The benchmarking tool tsmdiskperf.pl is available in the Blueprint configuration scripts package.
2. Run the benchmarking tool again and include more database disk subdirectories until throughput stabilizes or decreases.
3. Use the highest throughput value as the maximum to estimate database disk throughput during database backup operations to cloud object storage.
4. If the database disk throughput estimate is too low to achieve adequate throughput for database backup operations, reconfigure the database disks. For example, you can get greater random and sequential throughput by provisioning more physical disk volumes for database disk volume groups.