| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Website Support
Thread: Validation Running Behind? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 57
|
|
| Author |
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: 
|
Re outage announcement: http://www.worldcommunitygrid.org/forums/wcg/viewpostinthread?post=408639 Maybe the "various" word used in noted knreed post on the Jan.22 24 hour long outage [starting 03:00 UTC], was meant to be "varying" sizes. The idea is still to create WU's for sciences [where it's possible!], that are matched to the groups of hosts with different power [Limiting the sample to just for CPU, a centrino duo getting a WU with 2 HCC images, a I7-2600 getting WU's with 10 images or the former getting a FAAH job with 20 dockings and the latter with 100 dockings]. In all, when the average target run time of a science is 6 hours, it will be much closer to that target of 6 hours for all and not the sum of the slowest to the fastest and within runtimes ranging from 2 to 24 hours. Here's hoping I understand that approach and it indeed makes sense, then server hardware / memory capacity and network bandwith has to be set accordingly to that average. In WCG the challenge is to try to attract as much as possible the general public as well as institutions, companies etc.. to come and crunch so as to have a large number of projects that can benefit in an acceptable timeframe for a project. In such a context having always the right hardware/software infrastructure to satisfy everybody with a high availibility ratio, is an incredible challenge with a cost attached to it. Excellent job done by the tech team that makes this possible and thanks to IBM that finances this.  |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
The last post of knreed regarding the major database issues mentions as a possibility the repacking of WUs. It goes in the right direction. But knreed please if you repacking GPU WUs just tell us well in advance so that we can adapt the app_info.xml file to change the WU type number and avoid having idle times on the dedicated machines. We will give advance notice. There are a lot of strong reasons to do this since it would be great if we got into a situation where every processing resource could return at least 2-3 results per day. This would mean putting enough work in so that big power GPU's would only return a few results per day (but they would be big ones) while your Android cell phone would also be able to complete a couple while it is plugged in at night. It will also be good in that some of our volunteers have pretty solid PC's, but they don't use the PC's much. A mechanism that accounts for that and a project that has a wide range of job sizes available would allow that PC to also return work on a regular basis. Since we want to address this in a comprehensive solution, we will be working with Berkeley to make this a core feature in BOINC. Once that is done, we then hope to apply it to multiple of our science apps (where possible). But that will take some time and testing so it is a ways off yet. |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
The last post of knreed regarding the major database issues mentions as a possibility the repacking of WUs. It goes in the right direction. But knreed please if you repacking GPU WUs just tell us well in advance so that we can adapt the app_info.xml file to change the WU type number and avoid having idle times on the dedicated machines. If you update your boinc client to 7.044 which uses an app_config instead of an app_info you won't have to worry about updating your app_info every time the WU type changes. The only caveat is that you'll need to finish all the WUs in your cache before you make the switch or they will be discarded.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
See no reason why when the tasks are dynamically sized, the science application number would change... a non-worry to me, but could always be missing something. There will be advance notice for sure when app_info gets prohibited... when app_config is formally adopted. Could as well upgrade now and forget about what is looming, whilst you're away.
|
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
See no reason why when the tasks are dynamically sized, the science application number would change... a non-worry to me, but could always be missing something. This would only apply to people using an app_info to run multiple tasks on 1 GPU. Could as well upgrade now and forget about what is looming, whilst you're away. That was exactly my point. Since the app-config contains no WU type info there's no need to worry when upgrades change said info.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, but the images are still processed sequential and as long as the application number mask does not change, you don't have to change app_info. Not anticipating any change to the app, be it containing 1-2-4-6-8-10 for HCC-GPU. All other sciences for CPU are not a concern anyhow... we've seen many resizings on them in past. (too expand for those readers that do not fully understand the meaning of what's discussed)
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, but the images are still processed sequential and as long as the application number mask does not change, you don't have to change app_info. SekeRob, I'm confused.  What is the 'application number mask' for HCC7.08-GPU-WUs ?; ; andzgridPost#812 ; |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Et Al, It's Sunday, so please don't panic. All validators are suspended until the DB backup is complete which lastly measured took somewhat over 6 hours, ETC around 15:00 UTC. Your morning stats will look low!
And, this will repeat on Wednesday...hmmm, maybe knreed could have scheduled that 24 hours backup/restore outage on the 23rd instead of the 22nd. Could be they want to have a good secondary backup if things go against nature [touch wood it doesn't]. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Et Al, It's Sunday, so please don't panic. It took me a bit to realize what I was seeing. When I said "Thanks knreed, et al!" I meant "Thanks knreed and others [who helped]." Dictionary.com: et al |
||
|
|
twilyth
Master Cruncher US Joined: Mar 30, 2007 Post Count: 2130 Status: Offline Project Badges: 
|
When I saw that my daily total dropped from over 800k to under 300k I almost defecated a coprolite.
----------------------------------------  |
||
|
|
|