| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Retired Forums
Forum: UD Windows Agent Support [Read Only]
Thread: Computers unusable on HPF2 project |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 15
|
|
| Author |
|
|
davidhobbs
Senior Cruncher England Joined: Dec 30, 2004 Post Count: 151 Status: Offline Project Badges: 
|
I'm very disappointed to find that 15 of my machines that contributed to the HPF1 project are now unusable on HPF2. I can accept the need for more RAM, and I could obviously add more where needed. But I don't understand why a 450 or 500MHz processor can not do the same job as a 550. OK, it will take a little longer, but it will still get the job done.
As it is, you've lost half my computing power and I'm stuck with 15 redundant PC's in perfect working order. David. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello davidhobbs,
I do not know what the requirements for the next project will be yet. In the mean time, you might want to look at http://distributedcomputing.info/projects.html In particular, you might be interested in the i386 client for SIMAP: http://boinc.bio.wzw.tum.de/boincsimap/ Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The speed requirement isn't totally arbitrary. The aim is for each work unit to be completed in ten hours, or one week maximum. Very old computers simply can't keep up, and the architecture of the grid means that when a work unit is returned late, the work is wasted - it has already been done by someone else.
Also, there are environmental considerations. A modern computer has far better power management and consumption. More importantly, a modern computer can do the same work in less time, consuming far less power per work unit. Take an example: suppose you have one fast computer, and one old, slow one. Normally you run both during the day, even though you don't use them both. Instead, you could run the fast machine overnight just one night a week, and it will do more work in that night than the slow machine does all week. That way, you can leave the slow machine powered down, or donate it to a recycling scheme. |
||
|
|
davidhobbs
Senior Cruncher England Joined: Dec 30, 2004 Post Count: 151 Status: Offline Project Badges:
|
Well,
I think this is where I take my leave of WCG, which is a shame after donating over 29 years of computing effort to HPF1. ... but I can tell when I'm not wanted. David. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
That's sad to read dear DavidHobbs, but backward compatibility efforts have their limits..... I would not give up so quick because an answer is not to your liking......i did some sums and agree that there's time the fiat 500 is retired as the fuel efficiency of the fiat 600 is much greater.
----------------------------------------Many use BOINC to spread their total computing power across projects, so rather than putting all in one basket, why not continue to help here and put your other machines on less demanding grid calcs ? There are about 20/25 to choose from that can be managed and timeshared from a single BOINC manager AND with BOINCview, supplemental management can be handled in a jiffy. take care here or elsewhere
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 3 times, last edit by Sekerob at Jun 24, 2006 1:44:49 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Your effort is greatly appreciated.
However, pragmatism is important, too. If any of your machines will be more productive working on some other grid project, or on no project at all, then you should go ahead and do the right thing. I hope you have some faster machines to spare for WCG. The sad truth is that life sciences problems need some serious hardware, and despite the best efforts of the scientists and coders to break the problem down, there is a limit to how small each chunk can be. Check back in when new projects are launched - hopefully one of them will have less "up to the minute" requirements. Remember, it isn't that WCG doesn't want your contribution, it's just that WCG can't use all your computers at the moment. I think it is better that you are told about this up front, instead of just wasting your time for no purpose (as other, less ethical grid projects have done in the past). |
||
|
|
davidhobbs
Senior Cruncher England Joined: Dec 30, 2004 Post Count: 151 Status: Offline Project Badges:
|
OK then,
I'll leave my faster machines at your disposal and put the slower ones to work elsewhere for the time being. Because of the significant cost of powering 30-odd dedicated machines I actually only run them during the night on off-peak electricity, so I'm afraid even my faster machines take three times longer to return a result than you might expect. However, my three fastest machines are always on, so you do benefit from them 24/7. David. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi davidhobbs,
This is part of the reason that we are running BOINC. Any of our members using BOINC has a large variety of projects available, and with BOINCView, a large cluster of computers can be managed from one console. Keep on crunching! Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
dh, thanks for sticking around
 |
||
|
|
JeromeD
Advanced Cruncher Québec Joined: Nov 24, 2005 Post Count: 57 Status: Offline Project Badges: 
|
Well, 29 years is surely a good work for HPF1. All my teem united we just got a little over 3 years of crunching for now...Thanks for staying with us, with your fastest computers.
---------------------------------------- |
||
|
|
|