Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 475
|
![]() |
Author |
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Nushi,
----------------------------------------you may want to reverify your project options. I just changed to NRW to test and immediately got RICE work. Are you getting messages in your client log to say there is none available, sending other work?
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Silver for Rice, go to Gold.
![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Bah, I have my preferences set for NRW only but keep getting other stuff so I'm guessing there's not enough work and the fastest crunchers get the stuff that is there... I've only got 1 Rice WU in the last 24 hours on my quad-core machine. Not much chance of getting ruby if the end is closing in this fast.[snip] As Sekerob said, check your options... you can also create another Device Profile on the MyGrid page, under Device Manager->Device Profiles. Besides the default profile (which I leave with all tasks selected, plus the 'If there is no work ...' and 'Please opt me in ...' boxes checked), you can use the links there to 'Create A New BOINC Device Profile' for "home" "work" and "school." Use one of those 3 extra profiles to select only the NRW task, UNcheck the 2 'If there is no work ...' and 'Please opt me in ...' boxes (because DDDT-2 might start before NRW ends). Set your Advanced options below that, then click the Save button. You might consider increasing the 'cache' in the "Workunit Cache Settings" section while making the profile to, say, 0.5 to 1 day (I think the default is 0.2 days), so if WCG's servers go offline for 4 or 5 hours (accidents happen; hardware breaks; et cetera), your devices don't 'run dry'. Also consider that in BOINC the Advanced->Preferences setting on the 'network usage' tab, "Additional work buffer" will override that Device Profile setting. You can set each of your devices to use that profile on the Device Manager page by selecting the device name (click its link in the Device Names column), choosing the new Device Profile from the picklist, then the Save button. Repeat for each device. The next time BOINC updates with the WCG Project (usually happens when BOINC requests more work, but you can force it to update by selecting the World Community Grid project name on the Projects tab, then clicking the Update button), it will start using the new profile. You can confirm that on the Messages tab... e.g. it will change from saying 'Computer location: default' and tell you instead 'New computer location: home' (or whichever home/work/school Device Profile you created and used for RICE). If you participate in beta testing with your "default" profile, the new profile should automatically get added with a check mark on the Beta Testing page, but you might want to double-check that setting. Then you can 'for sure' force all your devices to run only RICE (NRW) work. There is a way you can actually name the profile RICE, but it requires a 'kludge' to get your devices to use that profile. Still, it IS possible to concentrate all devices on 1 task using the tools the WCG project gives us on the MyGrid page (it's just not accessible during stats updates, twice/day... which typically takes less than an hour each time). |
||
|
Nushi
Cruncher Joined: Feb 11, 2006 Post Count: 4 Status: Offline |
I know I can have many profiles, and I do (Default and home). At the moment they happen to be configured identically since I configured them just last weekend for Rice only.
Unchecking the "If there is no work..." on the Default profile seems to have done the trick. I set my buffer frm 0,25 days to 0,5 days (in the BOINC client, not the device profile) and updated WCG, got Rice :) Could the fairly low buffer have been the culprit? Whatever the reason, I'm back on track to getting my ruby, yay :) |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
[snip] Could the fairly low buffer have been the culprit? Whatever the reason, I'm back on track to getting my ruby, yay :) I rather think it was the "If there is no work available for my computer for the projects I have selected above, please send me work from another project." (SIC) box. I've seen my quads get 2 or 3 work units, then while those are still downloading BOINC requests more... which means there were not enough work units from the RICE task in the hopper to fulfill the first request, so it had to ask again. It seems to me that if that box was checked, it would have just given me some work units from 'another' task to go along with the RICE WUs. |
||
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3715 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
When there is some shortage of WUs for a given project requesters are prioritized as follows:
----------------------------------------1. single project with "work from other projects" not allowed 2. single project with "work from other projects" allowed 3. multiproject. If you really want/need a given project you must select option 1. But in that case you better keep watching your client often enough if you don't want it to run dry in case of severe shortage for this project. Cheers. Jean. |
||
|
rembertw
Senior Cruncher Belgium Joined: Nov 21, 2005 Post Count: 275 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
But in that case you better keep watching your client often enough if you don't want it to run dry in case of severe shortage for this project. Or if it's really enormously unbelievably important to you, select another boinc project, with resource share low (like 25 or so), and give WCG a large(r) resource share (like 1000 or so) in addition to the above hints. This prevents running dry. |
||
|
Mathilde2006
Senior Cruncher Germany Joined: Sep 30, 2006 Post Count: 269 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
But in that case you better keep watching your client often enough if you don't want it to run dry in case of severe shortage for this project. Or if it's really enormously unbelievably important to you, select another boinc project, with resource share low (like 25 or so), and give WCG a large(r) resource share (like 1000 or so) in addition to the above hints. This prevents running dry. Getting rice-WUs is hard at the moment. Asking for 400,000 seconds and I'm getting one or two WUs or even none. I thought, everyone has full Dengue caches. ![]() ![]() [Edit 1 times, last edit by Mathilde2006 at Feb 18, 2010 10:05:26 PM] |
||
|
rembertw
Senior Cruncher Belgium Joined: Nov 21, 2005 Post Count: 275 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Not everyone has DDDT2 "A"-type capable machines
![]() But I'm not crunching Rice either at the moment so I'm not guilty here ![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
After I've lost internet connection for a while, Boinic some times fails to find the server for an age. But if I restart Boinic, problem solved and I able to return and get new work units. Other than this I havent had any issues getting rice work units.
|
||
|
|
![]() |