| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: OpenPandemics - COVID-19 Project
Thread: Quota. |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
With that big drive to prepare for a tsunami of new arrivals, back office performance enhancements and all kind, a bit surprising they haven't done this now, just 6 lines of text, no functionality change at all. Yes they've said they want to overhaul the profiles, but then imagine the confusion for those who've been here since longer so guess the 'external' identifiers could be put in a hover box that pops up when mousing, both descriptions.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
One thing this project has done is bring to light a problem in the BOINC client. 7.16.1 in Ubuntu and 7.16.6 from the Fedora EPEL both ignore the cache setting and download work until it reaches the 1000 limit or the 70 per core, which ever occurs first. I now have 11,130 work units downloaded with a cache setting of 2 days. The eight thread machine with 560 will probably not be able to run them all. Will have to put that machine in a profile and use the WU limit to control work since the client is useless in this regard. Luckily, the other machines are big enough to handle the 1000 limit before the 7 day deadline for the most part. There might be 2 or 3 slower 24 thread machines that will be close.
|
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
One thing this project has done is bring to light a problem in the BOINC client. 7.16.1 in Ubuntu and 7.16.6 from the Fedora EPEL both ignore the cache setting and download work until it reaches the 1000 limit or the 70 per core, which ever occurs first. I now have 11,130 work units downloaded with a cache setting of 2 days. The eight thread machine with 560 will probably not be able to run them all. Will have to put that machine in a profile and use the WU limit to control work since the client is useless in this regard. Luckily, the other machines are big enough to handle the 1000 limit before the 7 day deadline for the most part. There might be 2 or 3 slower 24 thread machines that will be close. FWIW I run 7.16.31 on Ubuntu and don't have that issue. Can you upgrade?
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 |
Are you running from the gianfranco PPA? I was until I started having problems with that version and went back to the one in the standard Ubuntu repos. I think I'm going to run as is for a few days to let the scheduler stabilize on OPN, then adjust as necessary.
|
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
Are you running from the gianfranco PPA? I was until I started having problems with that version and went back to the one in the standard Ubuntu repos. I think I'm going to run as is for a few days to let the scheduler stabilize on OPN, then adjust as necessary. It's been a while but I'm pretty sure I got it from the Ubuntu repos. Doesn't make sense that they would have an older version and not something newer. I'm running 16.04 LTS if that helps.
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.
----------------------------------------[Edit 1 times, last edit by nanoprobe at May 15, 2020 3:31:13 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Are you running from the gianfranco PPA? I was until I started having problems with that version and went back to the one in the standard Ubuntu repos. I think I'm going to run as is for a few days to let the scheduler stabilize on OPN, then adjust as necessary. Gianfranco is/was actually the package maintainer of BOINC for Debian/Ubuntu. Hard to think his builds cause issues. Berkeley for the first time in quite a few years released it's own ubuntu build. Get it here https://boinc.berkeley.edu/dl/boinc_ubuntu_7.16.6_x86_64-pc-linux-gnu.sh |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I used the Gianfranco version for years but about 6 to 8 months ago strange things started happening with his version. His version was the first time I noticed the work fetch anomaly and then there were several updates where the version information printed in the BOINC log didn't match with the Ubuntu apt information. It just raised a bunch of questions for me and I went back to the standard repository version. Once I did, the work fetch was resolved. Upgraded to 7.16.1 with the 20.04 upgrade and work fetch problem returns. I thought I saw a work fetch issue reported to BOINC alpha about 2 weeks ago. Don't remember the details exactly. Will go back and look again. If this issue doesn't resolve, I'll try the Berkeley version. Thanks
----------------------------------------You don't notice the work fetch issue as quickly with WCG due to the work unit limit capability. I noticed it again once I started downloading OPN since it was set to unlimited. [Edit 1 times, last edit by Former Member at May 15, 2020 5:04:52 PM] |
||
|
|
adrianxw
Senior Cruncher Denmark Joined: Apr 13, 2008 Post Count: 196 Status: Offline Project Badges: 
|
>>> I don't believe that this should be the deciding factor to sway you towards or away from a specific project, though.
I stopped crunching here because of the actions of a moderator in the forums, not because of some marketting decision. |
||
|
|
|