On 12/20/06, Harakiri wrote:
I will study through the suggested paper, however
i actually read a different paper which suggested
using less messages, i would imagine that for arrays
of numbers lets say 100 Millions - the network
messages become the critical factor.
IMHO,
It depends completely
Thanks for your input,
--- Andreas Schäfer wrote:
> that even though this results in _many_ messages,
> the algorithms
> optimal runtime complexity will compensate for it.
>
> But benchmark your own ;-)
I will study through the suggested paper, however
i actually read a different paper which s
Hiho,
On 12:33 Tue 19 Dec , Harakiri wrote:
> i need some input on the parallelisation of quicksort.
> If anyone knows a more appropriate forum/list than
> this, please tell me =)
uhm, do you have access to a library? Get yourself a textbook on
algorithms. Knuth should be a good basis to star
Hello,
i need some input on the parallelisation of quicksort.
If anyone knows a more appropriate forum/list than
this, please tell me =)
I've written 2 different approaches to the quicksort
parallelisation.
1. The "master" server splits the array of numbers
into N seperate parts, where N is the
