Thank you, Ivan! I really didn't understand the example, but now it is much more clear to me!
Best wishes, Arthur 2013/4/7 Ivan Raikov <[email protected]> > By the way, I realized that the example in the MPI egg documentation may > not be all that clear, so I have updated it and added some comments. > This is a very simple example of a master/worker pattern. You are welcome > to improve on this with larger data sets or CPU-intensive operations. > > -Ivan > > > On Fri, Mar 29, 2013 at 11:44 PM, Arthur Maciel <[email protected]>wrote: > >> Ivan, thank you for the info. >> >> Indeed I don't program anything serious enough to need parallelizing. I >> am just curious about how far I can get with Chicken. I've seen it doesn't >> support native threads and I was a bit worried because I thought that >> 'production' languages for real world tasks should provide them. But now I >> see there is another way around with the MPI egg. >> >> I'm a constant beginner when talking to programming and I'm very curious >> about PL, especially Chicken Scheme, but almost never I get my hands dirty >> with code. >> >> I really understand what you mean by high level abstractions over basic >> MPI functionality, but I don't know how to do it. However, I'm quite >> excited to know Chicken and FPL in more depth and I would like to help on >> any useful task. When you intend to develop it more, please count on me. >> >> Best wishes, >> >> Arthur >> >> >> Em 29/03/2013 03:51, "Ivan Raikov" <[email protected]> escreveu: >> >> Hi Arthur, >>> >>> I use mpi for parallelizing some moderately-sized number crunching >>> tasks (kd-tree queries over trees of size ~1e5 points. I have primarily >>> used the scatter/gather collective operations for this. Of course, MPI is a >>> bit low-level for a functional programming language, and one day I hope to >>> develop some high level abstractions over the basic MPI functionality. If >>> you have some idea what kind of concurrent patterns you want to use in your >>> application, I can suggest which MPI functions may be of use to you. >>> >>> >>> -Ivan >>> >>> >>> >>> On Fri, Mar 29, 2013 at 9:49 AM, Arthur Maciel >>> <[email protected]>wrote: >>> >>>> Hello! >>>> >>>> I'm on: >>>> Linux Mint 3.5.0-17-generic #28-Ubuntu SMP Tue Oct 9 19:31:23 UTC 2012 >>>> x86_64 x86_64 x86_64 GNU/Linux >>>> Chicken Scheme Version 4.8.0.1 (stability/4.8.0) (rev 54c391c) >>>> linux-unix-gnu-x86-64 [ 64bit manyargs dload ptables ] >>>> compiled 2013-01-17 on aeryn.xorinia.dim (Darwin) >>>> >>>> libopenmpi-dev is installed and chicken-install compiled mpi perfectly. >>>> >>>> But when I run 'csi -e "(use srfi-4 mpi) (MPI:init)"' I get exactly >>>> the same error as the one in salmonella report at >>>> http://tests.call-cc.org/master/linux/x86/2013/03/28/salmonella-report/test/mpi.html >>>> >>>> I don't know how to make it work. Any help is appreciated. >>>> >>>> BTW, I've got to mpi from >>>> http://wiki.call-cc.org/man/4/faq#does-chicken-support-native-threads, >>>> but I've seen it is not used by any other egg. Is it really useful to >>>> parallel programming? >>>> >>>> Best wishes, >>>> Arthur >>>> >>>> >>>> >>>> _______________________________________________ >>>> Chicken-users mailing list >>>> [email protected] >>>> https://lists.nongnu.org/mailman/listinfo/chicken-users >>>> >>>> >>> >
_______________________________________________ Chicken-users mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/chicken-users
