Official 2011 Pentathlon Forum
Official 2011 Pentathlon Forum
OK, the team is registered, the suggestions for projects passed on and the old thread locked.
This is now the only thread to post on regarding the 2011 Pentathlon. Post anything you want, ideas and suggestions, but please don't start any other threads as it's easier to keep everything in one place, thanks! :D
This is now the only thread to post on regarding the 2011 Pentathlon. Post anything you want, ideas and suggestions, but please don't start any other threads as it's easier to keep everything in one place, thanks! :D
The first project for the 2011 Pentathlon has been announced, see below for details and maybe cache a few
The second BOINC Pentathlon starts with the project in the discipline World Community Grid:
Help Conquer Cancer
All credits in this subproject granted between
05/05/2011, 0.00 (UTC) and
05/10/2011, 0.00 (UTC) will be taken into the BOINC Pentathlon validation. Please note that pending credits cannot be taken into account! The statistics for this project can be found at
http://www.seti-germany.de/boinc_pentat ... llengeid=1.
We wish you lots of fun crunching the project!
Regarding choice of the project:
This subproject was proposed most often with 8 votes by the following teams: boinc.at, BOINC.Italy, boinc.sk, BOINC@Poland, Overclock.net, Planet 3DNow!, Russia, Team 2ch.
The other projects to complete the picture:
4 votes for Clean Energy Project (BOINC@Heidelberg, L’Alliance Francophone, Meisterkuehler.de, SETI.USA)
4 votes for Computing for Clean Water (BOINC Synergy, BOINCstats, Team China, UK BOINC Team)
4 votes for FightAIDS@Home (Electronic Sports League (ESL), SETI.Germany, Universe Examiners, USA)
4 votes for Help Cure Muscular Dystrophy (BOINC@Karlsruhe, Czech National Team, Rechenkraft.net, TitanesDC)
-
- Posts: 1630
- Joined: Mon Nov 06, 2006 12:00 am
GPU Project is PrimeGrid
They announced the GPU Project - PrimeGrid
http://www.seti-germany.de/blog/lang/en ... entathlon/
Timings for PG credits:
8 May 2011, 0.00 (UTC) to 13 May 2011, 0.00 (UTC)
Regards
Zy
http://www.seti-germany.de/blog/lang/en ... entathlon/
Timings for PG credits:
8 May 2011, 0.00 (UTC) to 13 May 2011, 0.00 (UTC)
Regards
Zy
-
- Posts: 1630
- Joined: Mon Nov 06, 2006 12:00 am
Same here.melter65 wrote:So now you see where the tactics come in, Zy
I don't have a GPU, so I will carry on crunching WCG wu's until the 10th May and then get ready for the 3rd project
Those of you with GPU's who crunch PG should change over on the 8th May and finish off your remaining WCG wu's before the 10th May.
I've just been looking at the Official Pentathlon site to see if any more projects have been announced. They haven't. But looking at the 'Stats' section I found that Project 3 will be Rosetta@Home and Project 4 will be QMC.
So an update to the schedule:
05/05/11 - 10/05/11: Help Conquer Cancer
08/05/11 - 13/05/11: Primegrid (GPU)
11/05/11 - 16/05/11: QMC
11/05/11 - 16/05/11: Rosetta@Home
EDIT: Post edited to show correct sequence of 3rd and 4th Projects. Dates are NOT a typo, both projects will run at the same time, so we need to know which project gives the best credits!
So an update to the schedule:
05/05/11 - 10/05/11: Help Conquer Cancer
08/05/11 - 13/05/11: Primegrid (GPU)
11/05/11 - 16/05/11: QMC
11/05/11 - 16/05/11: Rosetta@Home
EDIT: Post edited to show correct sequence of 3rd and 4th Projects. Dates are NOT a typo, both projects will run at the same time, so we need to know which project gives the best credits!
Last edited by melter65 on Wed May 04, 2011 11:10 pm, edited 1 time in total.
Sorry Zy, can't help you with that, I'm signed up to both projects but don't know a great deal about either!
The UK is UTC+1hr so I'd suggest starting crunching WCG about 23.30hrs (UK) because the results are going to land after the 00.00hrs (UTC)
Remember, all results RECEIVED after 00.00hrs (UTC) count
The 3rd and 4th Projects have been officially announced. As predicted they are QMC and Rosetta@Home. The post above has been edited to show correct sequence and dates.
The 3rd and 4th project will run concurrently so we need to know which project pays the better credits, QMC or Rosetta@Home.
The UK is UTC+1hr so I'd suggest starting crunching WCG about 23.30hrs (UK) because the results are going to land after the 00.00hrs (UTC)
Remember, all results RECEIVED after 00.00hrs (UTC) count
The 3rd and 4th Projects have been officially announced. As predicted they are QMC and Rosetta@Home. The post above has been edited to show correct sequence and dates.
The 3rd and 4th project will run concurrently so we need to know which project pays the better credits, QMC or Rosetta@Home.
No worries, understood. I had a ferret at Rosetta. It appears there are user options in the preferences file for different lengths of WU. Admins claim no advantage as such with any particular length, just user preference over length of WU they want to crunch. Appears that default is 3 hours, but I can see a small twist towards the end of the time period where a couple of one hour WUs would make the deadline, and a three hour one would not.Sorry Zy, can't help you with that, I'm signed up to both projects but don't know a great deal about either!
Still trying to wrap me head around QMC, looks straightforward, and appears like a WU is around 4+ hours for a 1090T - not 100% certain, but it appears pretty close to that.
Regards
Zy
Had a ferret re credits. Looks like QMC is around +/- 75% better paid than Rosetta.
I think its also a question of what team will crunch what. I suspect the bigger teams will go Rosetta because its more popular (as such).
As the Pentathalon goes by points per placing, not credits (as such). I reckon its probably best therefore to go with QMC.
However I am not experienced with these two, so will go with the majority on it.
Regards
Zy
I think its also a question of what team will crunch what. I suspect the bigger teams will go Rosetta because its more popular (as such).
As the Pentathalon goes by points per placing, not credits (as such). I reckon its probably best therefore to go with QMC.
However I am not experienced with these two, so will go with the majority on it.
Regards
Zy
I'm not sure if you can crunch cpu and gpu wu's together, I understand that the gpu wu's can be a little flakey if they don't get all the attention.
If you can finish the WCG wu's you have left while starting the PG gpu wu's then OK, but if you can't then concentrate on the PG wu's and abort any WCG wu's you have left over
The 5th and final Project has been announced. It's yoyo@Home and will run from 00.00hrs (UTC) 14/05/11 to 00.00hrs (UTC) 19/05/11
If you can finish the WCG wu's you have left while starting the PG gpu wu's then OK, but if you can't then concentrate on the PG wu's and abort any WCG wu's you have left over
The 5th and final Project has been announced. It's yoyo@Home and will run from 00.00hrs (UTC) 14/05/11 to 00.00hrs (UTC) 19/05/11
-
- Posts: 5602
- Joined: Sat Jun 23, 2007 1:00 am
That may change when the GPUs come out to play (although as to whether that change proves to be to our advantage or not, well, I guess we'll have to wait and see).melter65 wrote:48hrs into the 2011 Pentathlon and UBT are still 22nd out of 27 teams. We are well in front of the five teams behind us, but aren't making much headway towards the teams ahead of us.
-
- Posts: 5602
- Joined: Sat Jun 23, 2007 1:00 am
Well, a quick check on BOINC Stats shows that UBT is 26th in the world for QMC, and 19th for Rosetta, which suggests that Rosetta is our stronger project. Of course what really matters is how we compare to the other teams in the Pentathlon, and I can't really answer that one. :dontknow:Zydor wrote:Decision time ...
Which one do you want us all to go for, can't afford to be split between the two we are not a big enough Team for that, QMC or Rosetta ??
Regards
Zy
Hmm, the best thing I can suggest is that we start crunching both projects, see which gives us the best results, and then concentrate our efforts there. :scratch:
Looking at it from the BOINCStats angle makes sense. When I looked again with that in mind, what comes out is our "main" Crunchers who come into Team Events are not crunching either .... as such. Hardly anyone is doing QMC.Joshrandom wrote: .... Well, a quick check on BOINC Stats shows that UBT is 26th in the world for QMC, and 19th for Rosetta, which suggests that Rosetta is our stronger project. Of course what really matters is how we compare to the other teams in the Pentathlon, and I can't really answer that one....
If we took existing Cruncher rates as a "plus" - Rosetta wins hands down. We would gain more effect from those who pitch in and switch for a Competition period by building on what is at Rosetta already. With QMC we would be vertually starting from scratch (as such). It all hangs on where the big teams jump onto - a couple are bound to do both anyway
Decisions Decisions
Final Vote of the Zydor Jury Going to reverse my previous thoughts, and back Rosetta.
If the majority want QMC, I'll happily go with the vote though :thumbleft:
Regards
Zy
-
- Posts: 1655
- Joined: Wed Sep 19, 2007 1:00 am
Whatever you want it to be - go into preferences - there are the usual default/home/work/school sets. Open one up and at the bottom you will see a drop down box where you select your prefered WU length. Select the one you want, save, change the relevant PC to that preference set, job done.
(There is no advantage to short or long - credits scale well between sizes - just pick your prefered size)
Regards
Zy
(There is no advantage to short or long - credits scale well between sizes - just pick your prefered size)
Regards
Zy
Just had an enlightening experience over at Rosetta. I went there to get used to what was what, run a few quick ones to get into Rosetta before the start etc. Being a Rosetta newb I hit some buffers which may help some - no doubt second nature to Rosetta fans ...... doubtless not exhaustive, but just a newb experiences
First off, and a cracker ..... in preferences, "Target CPU run time" apparently does not mean what it says {*sigh*}. Best way I can describe it is a cut and paste from some helpful advice given to me by a Mod there :
- .... expectations of what a 1hr work unit is are not realistic for R@h..... when you do that, some of the other nice things such as accurate progress %, and consistently completing within such a limited timeframe go out the door. Each task must complete at least one model. For some tasks you will see a model every 5 minutes or so, for others, it can take several hours. So, not all tasks are going to complete within your one hour target, and that is normal and to be expected.
Bottom line - the advice is apparently to go for the default setting (three hours) and leave it at that to prevent problems. Just leaving it alone will default to three hour WUs. It would seem from looking at threads that there is a case for going to four hours .... maybe Rosetta experienced Team Members could chip in on the four hour bit.
- It would seem that BOINC gets more confused than it usually does if frequent changes are made to WU length, the advice is apparently settle on a size of 3-4 hours or above, and stick with it.
- Direct quote from their advice pages:
Progress Percent not advancing?
Rosetta recomputes the progress percent at the end of each model. The model number is shown in the graphic. As long as the "steps" are continuing to progress, it is working. Once it completes the model it is working on, it will recompute the progress. At that point the progress % will be determine by looking at the time it took to complete the first model, as compared to your WU runtime preference. If your WU runtime preference is low (<4hrs) you will frequently see the progress % jump from 1.xx to 100%, or into the 50% range. Basically, each different protein takes a different time to crunch a model. Some proteins will crunch for several hours to complete a single model. Others will crunch a model every 5 or 10 minutes. It is the nature of the science being done with Rosetta.
BOINC and Rosetta have measures in place that will abort work units that aren't running properly. So, in general, unless you see some specific advice to the contrary, you should NOT abort work units (WUs). {sic: no prizes for guessing what I did rofl}
- Model Percentages, awareness of below will avoid confusion:
"To completion" time is going UP!
A: This is normal. The time gets revised when the progress % is recomputed at the end of each model. So at the end of a model the time to completion will drop. Then during the crunching of the next model it will increase again.
- Q: I'm familiar with SETI and BOINC already, but what should I know about Rosetta? See this link: Rosetta for existing BOINC'rs
That'll do I guess else it ends up like War and Peace ..... use the link and follow on through as your fancy takes you. Questions to Rosetta experts, that aint me (!), and any of the latter reading this spot anything that would help us Rosetta newb's feal free to add to this - pretty please
Regards
Zy
First off, and a cracker ..... in preferences, "Target CPU run time" apparently does not mean what it says {*sigh*}. Best way I can describe it is a cut and paste from some helpful advice given to me by a Mod there :
- .... expectations of what a 1hr work unit is are not realistic for R@h..... when you do that, some of the other nice things such as accurate progress %, and consistently completing within such a limited timeframe go out the door. Each task must complete at least one model. For some tasks you will see a model every 5 minutes or so, for others, it can take several hours. So, not all tasks are going to complete within your one hour target, and that is normal and to be expected.
Bottom line - the advice is apparently to go for the default setting (three hours) and leave it at that to prevent problems. Just leaving it alone will default to three hour WUs. It would seem from looking at threads that there is a case for going to four hours .... maybe Rosetta experienced Team Members could chip in on the four hour bit.
- It would seem that BOINC gets more confused than it usually does if frequent changes are made to WU length, the advice is apparently settle on a size of 3-4 hours or above, and stick with it.
- Direct quote from their advice pages:
Progress Percent not advancing?
Rosetta recomputes the progress percent at the end of each model. The model number is shown in the graphic. As long as the "steps" are continuing to progress, it is working. Once it completes the model it is working on, it will recompute the progress. At that point the progress % will be determine by looking at the time it took to complete the first model, as compared to your WU runtime preference. If your WU runtime preference is low (<4hrs) you will frequently see the progress % jump from 1.xx to 100%, or into the 50% range. Basically, each different protein takes a different time to crunch a model. Some proteins will crunch for several hours to complete a single model. Others will crunch a model every 5 or 10 minutes. It is the nature of the science being done with Rosetta.
BOINC and Rosetta have measures in place that will abort work units that aren't running properly. So, in general, unless you see some specific advice to the contrary, you should NOT abort work units (WUs). {sic: no prizes for guessing what I did rofl}
- Model Percentages, awareness of below will avoid confusion:
"To completion" time is going UP!
A: This is normal. The time gets revised when the progress % is recomputed at the end of each model. So at the end of a model the time to completion will drop. Then during the crunching of the next model it will increase again.
- Q: I'm familiar with SETI and BOINC already, but what should I know about Rosetta? See this link: Rosetta for existing BOINC'rs
That'll do I guess else it ends up like War and Peace ..... use the link and follow on through as your fancy takes you. Questions to Rosetta experts, that aint me (!), and any of the latter reading this spot anything that would help us Rosetta newb's feal free to add to this - pretty please
Regards
Zy
-
- Posts: 1655
- Joined: Wed Sep 19, 2007 1:00 am
-
- Posts: 5602
- Joined: Sat Jun 23, 2007 1:00 am
It's been a while since I last crunched Yoyo, but after a quick refresher I can at least confirm that ECM does pay more than Euler on my Ubuntu system, not sure about the other sub-projects as yet though.Zydor wrote:Grateful for some words of wisdom from Yoyo fans.
Which sub project is best for the Pentathelon? I get the impression nosing around Yoyo that ECM seems top of the pops to crunch.
What would your recommendation be for a Yoyo newb with an eye to Pentathelon results?
Regards
Zy
Three days left to go, so let's have a round up on the scores:
1st Project: WCG - 22nd
2nd Project: PG - 7th
3rd Project: Rosetta - 17th (finishes tonight)
4th Project: QMC - 22nd (finishes tonight)
5th Project: yoyo - 19th (finishes Wednesday night)
Our overall position so far is 18th out of 27 teams!
1st Project: WCG - 22nd
2nd Project: PG - 7th
3rd Project: Rosetta - 17th (finishes tonight)
4th Project: QMC - 22nd (finishes tonight)
5th Project: yoyo - 19th (finishes Wednesday night)
Our overall position so far is 18th out of 27 teams!
Anyone reading this who has some spare Cores they would be prepared pitch into this for the last two days would be greatly appreciated. There are just a few of us holding our own at present at Yoyo, and the next Team up (BOINC@Heidelberg) is piling in with a Cruncher who has 3xi7 920s, 1x1090T & a i7 980 going full tilt ....
If we can get another 6 - 9 Cores or more contributing, we could get above them for the final rankings. Its that 7th Cavalry Moment
(Select the Muon WUs if the 7th Cavalry has no personal preference , good mix of short and long WUs in that preference setting, credits scale ok, no advantage short or long)
Regards
Zy
If we can get another 6 - 9 Cores or more contributing, we could get above them for the final rankings. Its that 7th Cavalry Moment
(Select the Muon WUs if the 7th Cavalry has no personal preference , good mix of short and long WUs in that preference setting, credits scale ok, no advantage short or long)
Regards
Zy
The Pentathlon is over for another year, phew!
The final results are:
1st Project: WCG - 22nd
2nd Project: PrimeGrid - 7th
3rd Project: Rosetta - 17th
4th Project: QMC - 22nd
5th Project: yoyo - 20th
Final Overall placing: 18th (out of 27 teams)
Needless to say, we were No.1 UK team!
Well Done to everyone who took part, whether in all the projects, or just the ones you could spare a core on. Thank You!! :salute:
The final results are:
1st Project: WCG - 22nd
2nd Project: PrimeGrid - 7th
3rd Project: Rosetta - 17th
4th Project: QMC - 22nd
5th Project: yoyo - 20th
Final Overall placing: 18th (out of 27 teams)
Needless to say, we were No.1 UK team!
Well Done to everyone who took part, whether in all the projects, or just the ones you could spare a core on. Thank You!! :salute: