|
Administrator
|
It is about how busy the disk device is.
For a diskgroup (such a raid arrays, volume group etc), it is not quite a very useful indicator, because it's basically showing how saturated "a" disk device is. So when you have multiple disk in the backend, it is just a percentage about how busy are those disks for serving at least one request. But those types of arrays or groups (i.e SSD or a RAID arrays) can do multiple I/Os at the same time.. So for RAID arrays and modern SSDs, this number does not reflect the performance limits. If you have a single device in the backend, then you can use this indicatior.. Higher percentage (for instance %100) -> higher saturation.
formal definition: Percentage of CPU time during which I/O requests were issued to the device
|
Locked
