Yann, If you find any good docs on dsage please let me know. I'll devote some of my spare time to dsage. It will be good to get it running in Grid.
Serge [EMAIL PROTECTED] пишет: > On 17 sep, 12:36, Serge Salamanka <[EMAIL PROTECTED]> wrote: >> Hi, >> >> I'm also interested in using dsage. >> I'm involved in international collaboration project developing Grid >> technologies (http://balticgrid.org/). >> I have one supercomputer at my disposal >> (http://supercomp.basnet.by/index_en.html) and I want to install sage on >> worker nodes. >> Yann, do you generate Monte Carlo data for LHC experiment ? >> >> I would be grateful for any links and any collaboration on developing >> dsage for Grid computing. > > No, it's Monte-Carlo for some optimization problem in imaging. The > parallelization is very primitive really : I distribute jobs to > workers. I just vary some parameters, then generate a new job with > those new parameters, and queue the job. > > I used to do that manually : I make a parameter file, scp it to a > node, run the program there. Of course, it's impossible to do that > now, since I now need to make thousands of variations. Also, I want > the result of a job to influence the next parameter variation. So > here's where Sage comes in, and since I use Python scripts mixed with > calls to external C programs, and am a Sage afficionado, dsage looks > perfect. The whole thing is not even running yet, because I don't > really master dsage, hence those questions. > > And it's easier to parallelize a queue of jobs than an individual > job... > > Yann > >> Serge A. Salamanka >> Laboratory of High Performance Systems >> United Institute of Informatics Problems >> National Academy of Sciences of Belarus >> mail-to: [EMAIL PROTECTED] >> >> [EMAIL PROTECTED] wrote: >> >>> On Sep 16, 9:37 pm, "William Stein" <[EMAIL PROTECTED]> wrote: >>>> On Tue, Sep 16, 2008 at 12:16 PM, Yann Le Du <[EMAIL PROTECTED]> wrote: >>>>> Hello, >>>>> I tried to email the person apprently responsible for dsage, Yi Qiang, >>>>> about this, to no avail, so I turn to the list. >>>>> I use sage, v. 3.1.1, and am trying to build an application (Monte Carlo >>>>> stuff) and use dsage to parallelize the code : very easy stuff, just do a >>>>> series of jobs, done normally in sequence on a single computer, in >>>>> parallel on many. >>>> Is it a bunch of computers all over on a network? Just curious what sort >>>> of "many computers" you have at your disposal. >>> Yes, many computers over a network, some dual and some quad cores, >>> with all >>> cores of nearly equal power. >>>>> So I fiddled around with dsage, managed to understand the basics, and I >>>>> find it very good, yet I have a few questions/remarks : >>>>> 1/ Why isn't there a clear, publicized, illustrated description of how to >>>>> use dsage ? I managed to make it work, but only after googling hard. >>>>> 2/ How can I send a job to a worker that will output intermediate values ? >>>>> I mean, say the job sent to a particular worker computes some value, and >>>>> that it takes 100 iterations, how can I output temporary values every 10 >>>>> iterations and have the server report those intermediate values ? >>>>> 3/ I noticed that workers can connect any time, really, and receive jobs >>>>> even if they connect to the server only after the server started some >>>>> sequence of jobs, which is cool. But I also noticed that if a worker gets >>>>> killed, then its job gets lost. Isn't it possible for the server to check >>>>> if a worker is alive, every once in a while, and if not requeue its job ? >>>>> 4/ What is the function I can use to check which worker did what, and if >>>>> it's alive, and what job got interrupted. >>>>> 5/ What test can I apply to a dsage job to see if it's finished ? Say a >>>>> job outputs a list, and I want to plot it, can I say something like "If >>>>> there is some output, plot it, otherwise wait." ? >>>>> 6/ If you have any notes, drafts, illustrating some of dsage >>>>> functionalities, I'd be more than happy to check them out. >>>>> Cheers, >>>>> -- >>>>> Yann Le Du >>>> -- >>>> William Stein >>>> Associate Professor of Mathematics >>>> University of Washingtonhttp://wstein.org > > > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-support URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
