Too many WU's
Anonymous
2 Mar 2005 13:37:23 UTC
Topic 12793
(moderation:
> Just set up another comp to crunch E@H. Pref's were set to connect every 1
> day. So why did it download 6 WU's?
> At minimum this will take 60+ hours and more like 72 hours (3 days). No prob
> if you don't mind leaving it on 24/7 - but I don't want to. Also in order to
> complete these WU's I now cannot resource share with another project - unless
> I do leave it on 24/7.
>
> If you want the general public to participate, you really need to make E@H
> user friendly. By that I mean that after starting the project, they only need
> to check progress once a day max. The average Joe does not want to be chasing
> deadlines every day.
>
2005-03-02 12:35:07 [normal ] OS version Microsoft Windows XP Home Edition, Service Pack 2, (05.01.2600.00)
2005-03-02 12:35:07 [normal ] Request [HOST#55418] Database [HOST#55418] Request [RPC#1] Database [RPC#0]
2005-03-02 12:35:07 [normal ] Processing request from [USER#17457] [HOST#55418] [IP XX.XX.124.50] [RPC#1] core client version 4.19
2005-03-02 12:35:07 [normal ] [HOST#55418] got request for 138489.794783 seconds of work; available disk 1.000000 GB
Your machine requested 138489.794783 seconds of work. That's about 1.6 days, or 40 hours of work. Hence the 5 WU you got. There was also an earlier request for a smaller amount of work.
Bruce
Too many WU's
)
> Another point of confusion is that BOINC tries to keep the queue
> between x and 2x days (4.1x clients) per project.
John, could you please point me to the part of the scheduler code where the factor of 2 appears? I don't see it. I see that when the client requests n seconds of work, the scheduler keeps sending WU until it has exceeded n seconds, then it stops. Where is the factor of 2?
Bruce
> > > Another point of
)
> > > Another point of confusion is that BOINC tries to keep the queue
> > > between x and 2x days (4.1x clients) per project.
> >
> > John, could you please point me to the part of the scheduler code where
> the
> > factor of 2 appears? I don't see it. I see that when the client
> requests n
> > seconds of work, the scheduler keeps sending WU until it has exceeded n
> > seconds, then it stops. Where is the factor of 2?
> >
> > Bruce
> >
> It's in the file boinc_public/client/cs_scheduler.c about 2/3 to 3/4 of the
> way down (sorry don't have line number handy). Opps on looking closer the
> times 2 line is commented out in the public branch too, I thought it was just
> commented out in the development branch.
>
> // determine work requests for each project
> // NOTE: don't need to divide by active_frac etc.;
> // the scheduler does that (see sched/sched_send.C)
> //
> p->work_request = max(0.0,
> //(2*work_min_period - estimated_time_to_starvation)
> (work_min_period - estimated_time_to_starvation)
> * ncpus
> );
John,
FWIW, the E@H scheduler and backend code is ALL from BOINC development CVS.
The relevant code is in the same place: sched/sched_send.C:
reply.wreq.seconds_to_fill -= wu_seconds_filled;
The right hand side is obtained from estimate_wallclock_duration().
As far as I can tell, a request for N seconds of work gets the smallest amount possible which is >= N.
Bruce