My idea: - Create an array of sub refs (why using sub refs - best practice if u need different subs on each element) - Execute each element found on the array as thread, and collect results into another array (use the ref subs array index as index on the results array) - loop the array and create the report.
As for using threads / forks / modules that does the same - I assume that all goes to kernel forks - so it doesn't matter to me. Searching CPAN I found the module from Nicholas Clark quite nice, less work, more readability, and also support shared / none share environments, also support thread locking (not that I see u need any of these). Thanks Chanan -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Yossi Itzkovich Sent: Monday, June 08, 2009 3:52 PM To: Perl in Israel Subject: Re: [Israel.pm] Parallel jobs Because I want to collect the results , send a nice report etc. -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Avishalom Shalit Sent: Monday, June 08, 2009 3:55 PM To: Perl in Israel Subject: Re: [Israel.pm] Parallel jobs why does it have to be in the same program , if they don't communicate much? i mean, why not run a script to invoke many instances of the program to the stations ? 2009/6/8 Yossi Itzkovich <[email protected]>: > Thanks, > > I need them basically to run (more or less unrelated ) tasks in parallel. > For example: Connect to dozens of stations and run there a long operation. I > don't want to wait long time by doing them sequentially. > > I prefer threads, but I wrote in the original mail why I can't. > > > Yossi > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of > Avishalom Shalit > Sent: Monday, June 08, 2009 3:01 PM > To: Perl in Israel > Subject: Re: [Israel.pm] Parallel jobs > > resending > > 2009/6/8 Avishalom Shalit <[email protected]> >> >> i had used pearl threads, >> it isn't too complicated and most things i needed were threadsafe. >> however , they are not lightweight at all, >> they are masked forks, and like forks, you really want to spawn them as >> early as possible . >> >> what do you need them for ? >> >> 2009/6/8 Yossi Itzkovich <[email protected]> >>> >>> Hi, >>> >>> I would like to ask how you people work with parallel jobs >>> (processes/threads) in Perl (>=5.8.4). >>> I know there are simple built in commands (open, fork etc.) but I am asking >>> about higher order managers. >>> Threads might be good - but not enabled by default in Perl build -so I am >>> limited (also not all modules are thread safe). >>> I have been using Parallel::Jobs for a while, and I read a little about >>> Parallel::Forker, but I wanted to know if there are better alternatives out >>> there. >>> >>> >>> Thanks for your help. >>> >>> Yossi >>> _______________________________________________ >>> Perl mailing list >>> [email protected] >>> http://mail.perl.org.il/mailman/listinfo/perl >> >> >> >> -- >> -- vish > > > > -- > -- vish > _______________________________________________ > Perl mailing list > [email protected] > http://mail.perl.org.il/mailman/listinfo/perl > _______________________________________________ > Perl mailing list > [email protected] > http://mail.perl.org.il/mailman/listinfo/perl > -- -- vish _______________________________________________ Perl mailing list [email protected] http://mail.perl.org.il/mailman/listinfo/perl _______________________________________________ Perl mailing list [email protected] http://mail.perl.org.il/mailman/listinfo/perl No virus found in this incoming message. Checked by AVG - www.avg.com Version: 8.5.339 / Virus Database: 270.12.56/2161 - Release Date: 06/07/09 17:53:00 _______________________________________________ Perl mailing list [email protected] http://mail.perl.org.il/mailman/listinfo/perl
