| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Discovering Dengue Drugs - Together - Phase 2 Forum
Thread: It's raining Dengue, Hallelujah! |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 1130
|
|
| Author |
|
|
I need a bath
Senior Cruncher USA Joined: Apr 12, 2007 Post Count: 347 Status: Offline Project Badges: 
|
There IS a limit 80 wu/core/day
---------------------------------------- |
||
|
|
AnRM
Advanced Cruncher Canada Joined: Nov 17, 2004 Post Count: 102 Status: Offline Project Badges: 
|
There IS a limit 80 wu/core/day The operative word is 'box' not cores....if that can't be done, maybe reducing the amount per core might be useful. The bottom line is to spread the WUs around a bit. There are legitimate reasons to download many WUs ie large cache but I wonder at times.....  |
||
|
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3715 Status: Offline Project Badges:
|
There IS a limit 80 wu/core/day The operative word is 'box' not cores....if that can't be done, maybe reducing the amount per core might be useful. Reducing it might be silly, not useful. In fact, currently, we (CAs and techs) keep watching it carefully because with new powerful machines and shorter WUs we are closer to this limit. Remember, the purpose of a computing grid is to get work done, not to spread WUs evenly with risks to have idle powerful machines. Also, once again, the reason for these shortages is that mass distribution has not started yet. A single normal full day for this project might be more than 100,000 results with so short WUs and currently the total results since the very beginning is still less than 50,000! |
||
|
|
AnRM
Advanced Cruncher Canada Joined: Nov 17, 2004 Post Count: 102 Status: Offline Project Badges:
|
Silly me....I stand corrected. I hope Greg L and others like him will wait around long enough to contribute.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Also, once again, the reason for these shortages is that mass distribution has not started yet. Is there anyone out there who already can (and will) mention when we can expect mass distribution? [Edit 1 times, last edit by Former Member at Mar 27, 2010 7:25:21 PM] |
||
|
|
gb009761
Master Cruncher Scotland Joined: Apr 6, 2005 Post Count: 2987 Status: Offline Project Badges:
|
Also, once again, the reason for these shortages is that mass distribution has not started yet. Is there anyone out there who already can (and will) mention when we can expect mass distribution? Probably not - through the very nature of the project. It was stated, right at the beginning, that this project will have periods when there's no work available - basically, due to the cascade of phases (with manual intervention in between).  |
||
|
|
mclaver
Veteran Cruncher Joined: Dec 19, 2005 Post Count: 566 Status: Offline Project Badges: 
|
There IS a limit 80 wu/core/day The operative word is 'box' not cores....if that can't be done, maybe reducing the amount per core might be useful. Reducing it might be silly, not useful. In fact, currently, we (CAs and techs) keep watching it carefully because with new powerful machines and shorter WUs we are closer to this limit. Remember, the purpose of a computing grid is to get work done, not to spread WUs evenly with risks to have idle powerful machines. Also, once again, the reason for these shortages is that mass distribution has not started yet. A single normal full day for this project might be more than 100,000 results with so short WUs and currently the total results since the very beginning is still less than 50,000! I agree with this. My I7-975 has peaked at 77 WUs. I have this machine dedecated to WCC 24x7, and it does nothing else (I build machines as a hobby, from parts, and dedicate them to WCC when I am done) Statistics Date Total Run Time (y:d:h:m:s) Points Generated Results Returned 3/27/10 0:004:03:09:40 15,917 16 3/26/10 0:006:14:24:53 26,538 31 3/25/10 0:009:08:27:42 35,957 48 3/24/10 0:006:09:05:47 25,237 39 3/23/10 0:008:23:06:00 35,225 53 3/22/10 0:012:02:36:34 46,978 77 With the availability of the I7-980 it probably will peak ovr 80, with the shorter WUs. It makes no sense to have such a powerful machine sit idle, and not do work. You should consider raising the limit, as I think it is too low for the next generation machines. - Mitch    |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
About the only useful thing WCG could do on their end to spread out the WU's a bit is put a limit on them like the Beta's have, ie- you are only allowed 1 WU per core or thread. For examle, on an i7 920, you could only have 8 WU's at a time. You can still have whatever length queue you desire but only 1 WU/core from that particular project.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
mclaver, the limit is 80/thread/day so an i7 975 would be allowed 80x8=640 WU's/day. I have a Gulftown running here and even with short WU like these DDDT2's it will only crunch through less then 300 WU/day.
----------------------------------------[Edit 2 times, last edit by Former Member at Mar 27, 2010 10:27:14 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I hope, that it will continue raining for a while. I just came home with several boxes and cartons... The content...? A case, a power supply, a mainboard, a CPU, a fan, RAM, etc. etc. So, please, let it rain, until I built all together, and installed the OS...  Lets hear what your building?  Well, it's nothing special... It's a Athlon64 X2 4400, with a W2k3 OS on it. It's a video-recording server for a friend of mine. But... it's running 24/7, so two more days a day of crunching for a better world. |
||
|
|
|