Em 05/12/2009, às 13:24, Ingo Weinhold <ingo_weinhold@xxxxxx> escreveu:
You seem to be under the incorrect assumption that we primarily want a fairscheduler. [...] If a fair scheduler does achieve that --great! I wouldn't ignore other algorithms, just because they aren't fair.
Not quite... Notice that what I said was a follow-up to what Mikhail wrote:
"I know about bit operations and if you have seen fair scheduler there is pow2 which is implemented without resorting to _factorOf array"
Where he mentions a weighting factor used on a fair scheduler that is calculated using pow. Were it a different approach I'd have no issues with that (other than the fact that I did try this approach outside of the fair algorithm context, and I didn't like the outcome. Anyway, he might succeed in doing it on a better way, so I didn't criticise *that*. Only the use of exponential decays inside a fair scheduler.
Cheers,A.