I am trying to understand the values I am seeing in vmstat -d. In particular the time field for reading and writing. The man page states:
FIELD DESCRIPTION FOR DISK MODE
total: Total reads completed successfully
merged: grouped reads (resulting in one I/O)
sectors: Sectors read successfully
ms: milliseconds spent reading
total: Total writes completed successfully
merged: grouped writes (resulting in one I/O)
sectors: Sectors written successfully
ms: milliseconds spent writing
cur: I/O in progress
s: seconds spent for I/O
When I look at the values in the "ms" field they appear to an absolute count of time spent reading or writing to a particular device. Is this correct?
I tried deriving the amount of time spent during the inspection interval by taking the difference between intervals and dividing by 1000. After doing this I find that it sometimes reports more time reading or writing to a device than the length of inspection interval.
Can anybody explain what I am doing wrong or why this happening?