is it possible to do "perfect" resource scheduling...

Question about which tools to use, bugs, the best way to implement a function, etc should go here. Don't forget to see if your question is answered in the wiki first! When in doubt post here.
Post Reply
d11in

is it possible to do "perfect" resource scheduling...

Post by d11in »

if the scheduler knows exactly what the process will do? like a process is represented by a list or a tree, the scheduler knows the computing depth of this process, "he" also know what the process will do...


best regards
common

RE:is it possible to do

Post by common »

No, that wouldn't be very reasonable.  The OS can make some guesses about what the process is doing though.  For example, during each run, massive changes in esp and ebp together, may indicate recursion, where the process needs more priority.  A massive change in just esp may mean, more data on the stack that wasn't there before, but may not necessarily mean more power would be necessary.  Also, you could look at eax, ebx, ecx, and edx, but since math would be performed in these registers, don't count on them alone.
d11in

RE:is it possible to do

Post by d11in »

don't look at register level for this. for example: your program is written in lisp, that's evaluated in primitive functions (one program step) and translated in a binary tree (list). now the scheduler knows the tree (that's what the program will do/ the scheduler knwows exactly when a program requires an io operation or something other). there are multiply programs (trees) which the scheduler analyses and choose the fittest program next.
Anton

RE:is it possible to do

Post by Anton »

Yes, it is possible.
My thoughts:The process it self probobly knows better, what will it do next, so it can tell the os, and then the os can use that info as it wishes. Ofcourse it would be harder to program the program. For instance, the program itself can tell the os, what type of operations are most common:i/o or computation. And the so can us that to tune the scheduler.
Anton.
Post Reply