Managing application I/O workloads using maximum IOPS settings

When multiple applications use a common storage subsystem, it is important to balance application I/O requests in a way that allows multiple applications to co-exist in a shared environment. You can address this need by setting a maximum threshold on the I/O operations per second (IOPS) for the volumes of an application. The volumes of an application are grouped to form an application volume group. The maximum IOPS limit determines the maximum number of I/Os processed per second collectively by all the volumes in an application volume group.

When an I/O request comes in from an application, it is serviced by the volumes in the group until the application volume group reaches the IOPS limit. When the group exceeds this limit for a specified time interval, further I/O requests on the group are queued. The queued I/Os are taken up on priority in the next time interval along with new I/O requests from the application.

You will want to consider the following factors when you set the maximum IOPS threshold:

Figure: Managing application I/O workloads illustrates the process.

Figure: Managing application I/O workloads

Managing application I/O workloads