Hi,
thank you very much Simon & Don, for the answers.
The latest head gives great results on parallel programs. All cores are
now always at work as I hoped for. So, too, thanks to everybody involved
in the multicore improvements -- they should come very handy. :-)
Thanks again,
Christian
* Don Stewart
choener:
Hi,
having tried the 6.10.2rc1 release candidate, I still find that "parMap rnf xs" on a list of thunks xs does not optimally use all available processors. With N the number of cores, I still see that each block of N thunks (say: x_1 and x_2) has to be calculated before (x3 and x4) will be started.
Would there be hope that compiling the latest head instead of 2009/03/14 (rc1) gives better results?
Yes, definitely. The HEAD implements all the `par` improvements described in the recent "multicore runtime" paper, as well as giving detailed runtime statistics on spark use.