Message boards : News : Status/Plans, 7th August 2012
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24527 - Posted: 7 Aug 2012, 15:18:31 UTC

First, many thanks for your continued support. From my/CERN side all has been running rather well and I am submerged by results.
I now need to take some time to analyse them. In particular to decide between the two methods of computing the beam - beam effect.
Then I shall probably submit several studies to do an intensity scan where I study the beam - beam effect depending on the size, and hence charge, of the accelerated bunch of particles.

At the same time, I must finish the documentation of the "user"
infrastructure so that my colleagues may easily use BOINC as they return from vacation. In addition I want to set up a dedicated "user" system "lxtrack" in order to provide disk space here and to try and keep up with the results as they are returned.

I have to look at the Deadline problem for 10**7 turn jobs.
I set a bound of 30 days for any WU....need to discuss with Igor is that is NOT what you see at home. Of course we really want a low bound to get results back quickly, but I also want to use older slower systems. We shall have to work out some sort of compromise. My attempt as 10**7 turns was probably a bit over the top, but I was keen to try it.

We hope/expect to produce a valid MAC executable this week. I also need to add some new "physics", new elements, to Sixtrack as provided by a colleague. (Also need to add modifications for "Collimation" but they are not relevant to BOINC.)
The next version should also support SSE4.1.

I was very pleasantly surprised to win an NVIDIA TESLA C2075.
The catch is that I have to use it and program it with OpenACC. There will doubtless be some hiccups intsalling the board and the necessary
(PGI) software. I shall in fact try my "Tomography" application which already runs in parallel using HPF or openMP. If that works I shall seriously consider a multi-threaded Sixtrack (using GPUs or not) by tracking many more particles in each Work Unit. Non-trivial but rather exciting. I am just at the ideas stage here, but.....it would of course use multiple threads on a multiple core PC as well. A dream?

Finally, I have to take time to publish my work on floating-point portability and reproducibility. I believe I might be the only person who gets identical bit for bit 0 ULP different results after many Gigaflops with 5 different Fortran compilers at different levels of optimisation.




ID: 24527 · Report as offensive     Reply Quote
Uffe F

Send message
Joined: 9 Jan 08
Posts: 66
Credit: 727,923
RAC: 0
Message 24529 - Posted: 7 Aug 2012, 19:32:58 UTC - in response to Message 24527.  

Thank you for keeping us all up to date. I know how long it can take to actually make the paper for submittion. Hope you get it accepted at a good journal, since I think that it is outstanding work on numerical identity after so much data processing.
ID: 24529 · Report as offensive     Reply Quote
[AF>FAH-Addict.net]toTOW

Send message
Joined: 9 Oct 10
Posts: 77
Credit: 3,671,357
RAC: 0
Message 24530 - Posted: 7 Aug 2012, 20:40:24 UTC

Great news !

Do you know how long there won't be work for us while you work on your tasks ?
ID: 24530 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24531 - Posted: 7 Aug 2012, 21:39:03 UTC - in response to Message 24530.  

only a few days i hope,
ID: 24531 · Report as offensive     Reply Quote
Profile Tom95134

Send message
Joined: 4 May 07
Posts: 250
Credit: 826,541
RAC: 0
Message 24532 - Posted: 8 Aug 2012, 4:43:02 UTC - in response to Message 24531.  

only a few days i hope,

I have 5 Tasks that are still "In Progress". The Remaining time on the longest one is just over 24 hours (this is about two-1/2 days calendar time for me since I shut down overnight). Should I just kill them or let them finish?
ID: 24532 · Report as offensive     Reply Quote
Uffe F

Send message
Joined: 9 Jan 08
Posts: 66
Credit: 727,923
RAC: 0
Message 24536 - Posted: 8 Aug 2012, 7:55:06 UTC - in response to Message 24532.  
Last modified: 8 Aug 2012, 7:55:32 UTC

only a few days i hope,

I have 5 Tasks that are still "In Progress". The Remaining time on the longest one is just over 24 hours (this is about two-1/2 days calendar time for me since I shut down overnight). Should I just kill them or let them finish?


No, you should keep crunching them. He is not shutting down the servers, so he still needs the results. He is just not making any new work for the next few days.
ID: 24536 · Report as offensive     Reply Quote
Phil
Avatar

Send message
Joined: 26 Jul 05
Posts: 63
Credit: 4,083,755
RAC: 0
Message 24541 - Posted: 8 Aug 2012, 20:54:22 UTC - in response to Message 24527.  

Thanks for the update.
It is gratifying to know that the new executables coped will with the huge flood of jobs so well.
ID: 24541 · Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 24 Oct 04
Posts: 1114
Credit: 49,501,728
RAC: 4,157
Message 24544 - Posted: 9 Aug 2012, 5:50:01 UTC

Eric,

I have a task that is half finished and it has about 25hrs left which would be about 12hrs past deadline and then 2 more tasks like that on this one host.

And about another dozen on some of my other hosts........should we still run these even though they will be past deadline or abort?

Several of these will take over 50 hours each.





Volunteer Mad Scientist For Life
ID: 24544 · Report as offensive     Reply Quote
metalius
Avatar

Send message
Joined: 3 Oct 06
Posts: 101
Credit: 8,985,206
RAC: 43
Message 24545 - Posted: 9 Aug 2012, 6:49:03 UTC - in response to Message 24527.  
Last modified: 9 Aug 2012, 6:49:39 UTC

I set a bound of 30 days for any WU....need to discuss with Igor is that is NOT what you see at home. Of course we really want a low bound to get results back quickly, but I also want to use older slower systems. We shall have to work out some sort of compromise.

First, it is very nice to see, You are following basic principle of BOINC - everyone is welcome and everyone can participate!!!
The parent-veteran project SETI@home is giving very good example. SETI@home is using different deadlines (from 1 week to 2-3 months) according to the size of tasks.
IMHO, such solution is the best compromise.
ID: 24545 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24546 - Posted: 9 Aug 2012, 7:16:19 UTC - in response to Message 24532.  

Please let them run if OK with you. I hope they don't get
killed... Thanks. Eric.
ID: 24546 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24547 - Posted: 9 Aug 2012, 7:17:25 UTC - in response to Message 24545.  

thnaks; and indeed it was stupidity/oversight not to
have changed the deadline. Eric.
ID: 24547 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24548 - Posted: 9 Aug 2012, 7:19:45 UTC - in response to Message 24544.  

Well; I am farid they may get killed at deadline.
If you want to save resources you could kill them
yourself. I have a re-submit procedure BUT
only when I have fixed the deadline. Eric.
ID: 24548 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24549 - Posted: 9 Aug 2012, 7:21:05 UTC - in response to Message 24530.  

I hope to prepare new work over the weekend. Eric.
ID: 24549 · Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 24 Oct 04
Posts: 1114
Credit: 49,501,728
RAC: 4,157
Message 24550 - Posted: 9 Aug 2012, 8:46:05 UTC - in response to Message 24548.  

Well; I am farid they may get killed at deadline.
If you want to save resources you could kill them
yourself. I have a re-submit procedure BUT
only when I have fixed the deadline. Eric.




Ok I just deleted the 2 long tasks I had on this host and will just check the other tasks I have on the other hosts since a few are not due until the 13th and some are the shorter versions.

Hard for me to kill the long one I have running on this host as I type since it has been running the task for 30 hours but I can't think of a way to get the remaining 20+ hours finished in just 4 hours

But as always I will be here when you get this all fixed this weekend and get back to work here Eric.

Good luck



Volunteer Mad Scientist For Life
ID: 24550 · Report as offensive     Reply Quote
[AF>FAH-Addict.net]toTOW

Send message
Joined: 9 Oct 10
Posts: 77
Credit: 3,671,357
RAC: 0
Message 24551 - Posted: 9 Aug 2012, 10:42:32 UTC - in response to Message 24549.  

I hope to prepare new work over the weekend. Eric.

Nice :)

Will you include some new features that you talked about, or aren't they ready yet ?
ID: 24551 · Report as offensive     Reply Quote
Christoph

Send message
Joined: 25 Aug 05
Posts: 69
Credit: 306,627
RAC: 0
Message 24554 - Posted: 9 Aug 2012, 20:18:08 UTC

In case that you get this problem again in the future, I think somewhere I read that it is also possible to extend the deadline for already sent out tasks.
Propably it was at SETI@home after some of their unplanned-couple-of-days-long outages.
They extended the dealines so that it was possible to report the already overdue tasks.

Sorry that I didn't remembered earlier about that.
Christoph
ID: 24554 · Report as offensive     Reply Quote
Richard Haselgrove

Send message
Joined: 27 Oct 07
Posts: 186
Credit: 3,297,640
RAC: 0
Message 24555 - Posted: 9 Aug 2012, 22:42:32 UTC - in response to Message 24549.  

I hope to prepare new work over the weekend. Eric.

Looks like the weekend arrives early in your part of the world :)
ID: 24555 · Report as offensive     Reply Quote
[AF>FAH-Addict.net]toTOW

Send message
Joined: 9 Oct 10
Posts: 77
Credit: 3,671,357
RAC: 0
Message 24556 - Posted: 10 Aug 2012, 9:17:19 UTC

It was a quick batch ... I'm not sure this is the one that Eric was referring to ...
ID: 24556 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24557 - Posted: 10 Aug 2012, 9:43:57 UTC - in response to Message 24551.  

Probably stick with what we have. I need to do more
testing and consult wuth Igor. Eric.
ID: 24557 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 857
Credit: 1,619,050
RAC: 0
Message 24558 - Posted: 10 Aug 2012, 9:48:38 UTC - in response to Message 24554.  

Thanks; I suspect something is going on here as I mentioned before.
I specify a deadline of about 2,400,000 seconds, about 30 days I
think but this os not what you see. We also try to get a study finished
without waiting forever. Interestingly enough I have got back 90% of
my results for 10 million turns, but they may not have run to completion.
Eric.
ID: 24558 · Report as offensive     Reply Quote
1 · 2 · Next

Message boards : News : Status/Plans, 7th August 2012


©2024 CERN