Question - How Large Is It Preferred To Configure The Data Volumes Of A Sap Maxdb Database?
Answer -
The optimum use of the I/O system is critical for I/O performance. Thus, it becomes beneficial to distribute the volumes evenly across the available I/O channels.
The parallelism of the I/O gets affected by the number of data volumes.
Windows:
The asynchronous I/O of the operating system is used on Windows.
UNIX:
On UNIX the number of configured I/O threads determines the parallelism with which the SAP MaxDB/liveCache database transfers the I/O requests to the operating system.
o SAP MaxDB version lower than Version 7.7
The number of volumes * number of I/O threads for each volume (_IOPROCS_PER_DEV), gives the number of I/O threads.
o SAP MaxDB Version 7.7 or higher
Volumes * (total of low/med/high queues for each volume), gives the number of I/O threads. But it can get limited by the database parameter MaxIOPoolWorkers.
The number of threads gets increased by a number of threads that was configured too high. This results in reaching the limits of the operating system resources.
It is recommended to use the following formula to determine the size of SAP MaxDB data volumes: 'square root of the system size in GB, rounded up'.
Examples:
10 GB: 4 data volumes
50 GB: 8 data volumes
100 GB: 10 data volumes
200 GB: 15 data volumes
500 GB: 23 data volumes
1 TB: 32 data volumes