| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: New Beta Test – May 29, 2019 [Issues Thread] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 315
|
|
| Author |
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 858 Status: Offline Project Badges: 
|
I would hope that when this project goes live it will have similar warnings about the type of systems needed to run the tasks within reasonable times (along with a recommendation to use LAIM to help counter the restart issue.) And I think it will need an explicit opt-in project, so no "automatic opt in" when it goes live... This project will be opt-in only when launched and will call out the caveats on the project selection page. Also it is likely when launched the default will be set to only allow one task per host at a time. Thanks, armstrdj Thanks for repeating this - I knew I'd seen it somewhere up-thread but I wanted to reply to NixChix and then went to find it - I'll edit that link in there too "for completeness"... And that "1 per host" restriction seems like a good idea - I wish CPDN would do something similar for their much longer jobs! Once again, thanks, both for this post and all the hard work that has gone into getting this ready for production. Cheers - Al. |
||
|
|
hchc
Veteran Cruncher USA Joined: Aug 15, 2006 Post Count: 746 Status: Offline Project Badges:
|
Ouch, 1 task per host makes it tough for badge hungers who want to crunch ARP1 100% of the time 24/7 who reliably return the tasks.
----------------------------------------Can we have a middle ground and limit it to 1 task per thread like in the Beta? This would at least let us keep all cores/threads topped off.
|
||
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges: 
|
One thing I can say for sure though is that despite the apparent size of the calculations being performed, these don't seem to reduce overall machine throughput in the way MIP1 does  That reminds me that when I ran the ARP1 along with the MIP1, there seemed to be an interaction such that the MIP1 time lengthened. And it could be the case that the ARP1 time lengthened also, but I didn't get enough to do much of a test. But it will be on my to-do list when they are released. |
||
|
|
armstrdj
Former World Community Grid Tech Joined: Oct 21, 2004 Post Count: 695 Status: Offline Project Badges: 
|
Ouch, 1 task per host makes it tough for badge hungers who want to crunch ARP1 100% of the time 24/7 who reliably return the tasks. Can we have a middle ground and limit it to 1 task per thread like in the Beta? This would at least let us keep all cores/threads topped off. One task per host would be the default but not the limit. A volunteer could set it to whatever they like. Thanks, armstrdj |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
One task per host would be the default but not the limit. A volunteer could set it to whatever they like. Thanks, armstrdj Like CEP2 was. Any time frame on if/when there will more betas released for either project. Will be leaving town tomorrow for a few days and want to have everything set up in advance. Thanks.
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.
  |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1258 Status: Offline Project Badges: 
|
This application is heavy in floating point calculations and without some optimization limitations we would not get binary equivalence due to rounding differences even on different current generation processors. Thanks, armstrdj In terms of the application having a heavy floating point as it's too early to determine whether it run faster on Intel or AMD under Windows 10?  |
||
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges:
|
In terms of the application having a heavy floating point as it's too early to determine whether it run faster on Intel or AMD under Windows 10? I think this pretty much answers it. Linux is faster. https://www.worldcommunitygrid.org/forums/wcg...ad,41610_offset,40#610608 |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
*chirp*
----------------------------------------
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 |
May not hear much the next couple of weeks as this is a shortened holiday week and the BOINC Workshop will be going on in Chicago the week of the 8th.
|
||
|
|
Seoulpowergrid
Veteran Cruncher Joined: Apr 12, 2013 Post Count: 815 Status: Offline Project Badges: 
|
I competently forgot about the holiday in the US. Thanks for the reminder on both of those events.
---------------------------------------- |
||
|
|
|