looking for multicore scheduler flaws

[radj]

hello, guys

has there been any evaluation on the new multicore scheduling support that the new kernel provides?

havent been able to find any.

im asking this coz its the first step to a thesis we are working on. Improving Linux Multicore Scheduler

the first step we hav to do is find, if there are any, existing flaws of the multicore scheduler. and we'l try to improve the flaws.

so far, the evaluations i've found are for the unicore processor, the flaw with the lag on response time? i bet it's not a problem anymore with multicore processing.

any problems will do.

thanks.

if you have suggestions on how we should go about our thesis, they are totally welcome.

[burntfuse]

Sorry, I don't know anything about this, but it sounds like an interesting project. I hope you're planning to get your improvements included in the mainstream kernel?

[radj]

i chose this kind of thesis coz i know it's helpful. i'd like to contribute to the actual kernel too. that is if the improvments are remarkable. ehehehhe.