mark somers schrieb:
>> We derived the start source code from Leiden Classic, which is in trac.
>> But we changed a lot.
>>     
> So, what did you change then? Is it worth while to update trac then? The 
> functionality you describe seems to be still the same...
>
> m.
>   
We changed it to fit our applications. I think it is to specific to our
needs than to share it in trac. Main differences:
- memory-, runtime estimations are not done by putting the values into
the job submission formular. These values are estimated by our work unit
generator
- we check the commandline parameter which you can add in the formular
specific to our applications.
- by default job submission is forbidden for all users and only allowed
for users which are listed in q_users, so we don't use q_restricted_apps
anymore
- solved the  issue to get rid of finished user jobs. User hav to
download and delete finished jobs to start new ones. I think this wasn't
in trac.

yoyo

Rechenkraft.net e.V. <http://www.rechenkraft.net> - Verein zur Förderung
von Bildung, Forschung und Wissenschaft durch Einsatz vernetzter Computer
weitere interessante Projekte
<http://www.rechenkraft.net/wiki/index.php?title=Projekt%C3%BCbersicht>und
Hilfe <http://www.rechenkraft.net/phpBB/viewforum.php?f=19>auf unserer
Webseite www.Rechenkraft.net <http://www.rechenkraft.net> und im Chat
<irc://irc.euirc.net/rechenkraft>
------------------------------------------------------------------------
Rechenkraft.net e.V. <http://www.rechenkraft.net/>  - Non-profit
association for the promotion of education, research and science through
the use of networked computers
other interesting projects
<http://www.rechenkraft.net/wiki/index.php?title=Projekt%C3%BCbersicht>and
help <http://www.rechenkraft.net/phpBB/viewforum.php?f=19> on our
website www.Rechenkraft.net <http://www.rechenkraft.net> and on IRC
<irc://irc.euirc.net/rechenkraft>

_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.

Reply via email to