Oliver Smith wrote: > Download: http://www.kfs.org/async/worker/ > Documentation: http://www.kfs.org/async/worker/manual/
We should somehow store pointers to 0MQ-related projects like this one on zeromq.org website, otherwise they'll be hard to find and hard to use. Anyone any idea of how it should be done? A simple "links" section? > It combines concepts from OpenMP and Intel's TBB. I am not the expert on either of the two, but the idea of making 0MQ infrastructure look more friendly to OpenMP/TBB developers looks like an interesting idea. If you are interested in discussing it, comparing the OpenMP/TBB/0MQ approaches, benchmarks etc. you can possibly write a blog about it to post on zeromq.org. > This is a somewhat weak example because the work being done by the > worker is so trivial, but even so on a virtual quad-core machine > building with -O0 I see a 35-40% reduction in processing time. Wrker being trivial, the large reduction in processing time is even more impressive. > I ran a second test with a ridiculously complicated formula involving > int-to-float, float-to-double, double-to-bool, sin, cos, abs, modulo etc > operations, and benchmarked running the array in serial vs parallel. > > Serial: 6784 miliseconds > Paralle: 1401 miliseconds. > > And that's without any effort to take things like cache alignment, cache > lines etc into account, or any effort to tweak the number of tasks per > workload. Nice! > > I hope some of you will find this useful. > > If you want to provide feedback or questions, either follow up here, to > my email address or on the blog post linked from the manual page. > > Happy threading :) Martin _______________________________________________ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
