Hi Rick! It depends on what should be the goal of measurement. IOPs is only one of the parameters that determine the quality of IO subsystem. The total amount of IOPs depends on the storage used, on the physical interfaces of the storage/host etc. More interesting, according to me, is the number of IO operations that are waiting for processing. When the queue is small or empty, the IO is (from point of view of the virtual machine) okay. The problem usually starts when the queue size will increase. Then it is important to measure the IOPs, but also the amount of data transmitted.
Keep Up with TechRepublic