How the scheduling algorithm determines the type of process (I / O, processor binding)
I read scheduling algorithms and wonder how the scheduling algorithm determines if a process is cpu bound
or io bound
and makes a decision accordingly? For example, in an implementation, multilevel feedback queue
he mentions that he gives preference to a short problem related to i / o. My question is, how will he determine this?
source to share
In Linux 2.6, a process is considered interactive if its dynamic priority is greater than its static priority.
Linux updates the priority of each process dynamically according to its average latency . (Without going into exact calculations and functions).
We expect the process cpu bound
to have a low average latency. On the contrary, we expect the process I/O bound
to have high latency.
source to share