The tasks that use their full time slice might get their priority lowered. This will make I/O bound tasks more likely to run on the CPU, hence, more tasks will be run in a time window. However, we need to ensure not to starve CPU-bound tasks too much if there are lots of I/O bound tasks.
IIRC, from the OSTEP book, we might increase the priority of the tasks that did not run after a time slice. But this ofc comes with the cost of operation. Anyways, there is a good amount of thing to think here but the OS itself is not closely mature enough to be able to test things like this. So for now, just dropping this so that I don't forget.
The tasks that use their full time slice might get their priority lowered. This will make I/O bound tasks more likely to run on the CPU, hence, more tasks will be run in a time window. However, we need to ensure not to starve CPU-bound tasks too much if there are lots of I/O bound tasks.
IIRC, from the OSTEP book, we might increase the priority of the tasks that did not run after a time slice. But this ofc comes with the cost of operation. Anyways, there is a good amount of thing to think here but the OS itself is not closely mature enough to be able to test things like this. So for now, just dropping this so that I don't forget.