| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: Building the 400 Teraflop Machine... a Work In [Permanent] Progress |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 101
|
|
| Author |
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
But for CEP2 there really should not be and that element is directly going to Harvard. The DDDT2 are sizable but nothing new is coming out in downloads the past few days but repair work plus this science moderated traffic and the C4CW has a large but one time big download and after actually is highly efficient since its only a seed signal that is set to activate additional tasks on a host. At any rate, as knreed posted some days ago about the upgrades and the work to put a highly up-scalable grid in place, handling a million tasks/results daily, what's there left. He joked on doubling the volunteer contributions: "Bring it on".
----------------------------------------
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Sep 17, 2010 4:23:11 PM] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
Check out https://secure.worldcommunitygrid.org/help/vi...?shortName=minimumreq#413
You reported as complete almost 5000 results for cep2 in the past 36 hours. At 40MB each that is about 195GB of data. You still have 2500 'in progress' and I suspect that many of those are still trying to upload. Harvard is working at improving their ability to handle the volume as we get ready to launch windows and I think you helpfully gave them a bit a stress test yesterday. However, it might be helpful if you deselect cep2 from your profile for a bit while they work on adjusting things. |
||
|
|
anhhai
Veteran Cruncher Joined: Mar 22, 2005 Post Count: 839 Status: Offline Project Badges: 
|
I was wondering why all of a sudden MrKermit dropped from being #1 cruncher. It was simplify due to the fact that he did too much crunching and wound up accidentally crunching on harvard's servers
---------------------------------------- MrKermit, thx for contributing your Super Duper computer to the effort when it is not being used.  |
||
|
|
Dataman
Ace Cruncher Joined: Nov 16, 2004 Post Count: 4865 Status: Offline Project Badges: 
|
I think you helpfully gave them a bit a stress test yesterday. However, it might be helpful if you deselect cep2 from your profile for a bit while they work on adjusting things. This made my day!! Unexpected?  Bring it on ...........   [Edit 1 times, last edit by Dataman at Sep 17, 2010 8:17:48 PM] |
||
|
|
RaymondFO
Veteran Cruncher USA Joined: Nov 30, 2004 Post Count: 561 Status: Offline Project Badges: 
|
Sekerob,
----------------------------------------Love your linux penguin. Do you mind if I use this as well? Ray [Edit 2 times, last edit by RaymondFO at Sep 17, 2010 8:27:37 PM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Plz go ahead. The smiley provider may not be happy me using it in my sig showing in almost 19k posts, so I'll probably shadow it to my Chart hoster which has unlimited bandwidth use till July 2012
----------------------------------------Oh, forgot the price: crunch crunch Crunch Crunch Carunch, at WCG enjoy
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
As for MrKermit's experience, I think that henceforth we can coin the famous Muppet Show phrase to have a new meaning at WCG:
----------------------------------------"The: It's not easy to be Green effect". 
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
MrKermit
Advanced Cruncher Joined: Jun 13, 2009 Post Count: 95 Status: Offline Project Badges:
|
Exactly what I was looking for ... thanks! You reported as complete almost 5000 results for cep2 in the past 36 hours. At 40MB each that is about 195GB of data. You still have 2500 'in progress' and I suspect that many of those are still trying to upload. Definitely, I see a lot of them failing still. Harvard is working at improving their ability to handle the volume as we get ready to launch windows and I think you helpfully gave them a bit a stress test yesterday. However, it might be helpful if you deselect cep2 from your profile for a bit while they work on adjusting things. Deselected for now. It's really not that much data though. I have my nodes throttled on upload and download bandwidth and I suspect they may be timing out because they are going too slow or something. My favorite part of the chart above is the Hours per task plummeting. It really helps me perceive my systems as efficient relatively speaking. Cheers! MrKermit ----------------------------------------  |
||
|
|
anhhai
Veteran Cruncher Joined: Mar 22, 2005 Post Count: 839 Status: Offline Project Badges:
|
Check out https://secure.worldcommunitygrid.org/help/vi...?shortName=minimumreq#413 You reported as complete almost 5000 results for cep2 in the past 36 hours. At 40MB each that is about 195GB of data. You still have 2500 'in progress' and I suspect that many of those are still trying to upload. Harvard is working at improving their ability to handle the volume as we get ready to launch windows and I think you helpfully gave them a bit a stress test yesterday. However, it might be helpful if you deselect cep2 from your profile for a bit while they work on adjusting things. I am wondering if this is only meant to help MrKermit, because I really don't understand how MrKermit deselecting cep2 will help the situation for harvard. Harvard should get roughly the same number of WU sent back to it if not more, this is because of how the WU are distribute (or at least based on my limited knowledge). According Knreed and Sekerob, the way it works is that WCG has some feeders that assign WUs and in the feeder there are X number of HCC WUs, Y number of HFCC WUs, Z number of CEP2 WUs, ... and the feeder doesn't get refilled until it is totally empty. So this means that for every X number of HCC WU gets sent/crunched then Z number of CEP2 WU must also get done. So if MrKermit isn't crunching CEP2 WUs anymore, all this means is that the rest of us will have more CEP2 WUs to crunch (if we selected to do this project). The only solution is to lower the weight of CEP2 WU while MrKermit is crunching. Of course I may be totally of the mark here. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
CEP in running is Beta and uncovering many issues. but next week it will start again.
Have a nice weekend. |
||
|
|
|