| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Teams
Thread: SNURK and friends - Hop on to say hi |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 1141
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I’ve looked again at machines today, none of them seem to have more than 5 days worth of tasks so Wednesday will be my cutoff point, barring some kind of disaster we should be eligible for week 3
 |
||
|
|
DCS1955
Veteran Cruncher USA Joined: May 24, 2016 Post Count: 668 Status: Offline Project Badges: 
|
I’ve looked again at machines today, none of them seem to have more than 5 days worth of tasks so Wednesday will be my cutoff point, barring some kind of disaster we should be eligible for week 3 Would that be Wed Oct 2?   |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You don't want to bunker now unless you plan on releasing them before Sunday. This week, just let them run and report. On Sunday after 00:00UTC, that's when you load up and turn off networking. Beware, you need to bunker the work that has a 10 day deadline(MIP ZIKA TB) or you may lose some work the following Sunday/Monday. The reason you are probably being limited to 5 days is because you are reaching the WCG imposed limit of 70 per thread. For an 8 core machine, that is 560 WUs (however long it takes your machine to run that). To get around that, you have to change ncpus in cc_config.xml to some higher value than your current number of cores. Here's the rub with that, you will also need to add (if you don't already have one) an app_config.xml file in your <BOINC DATA>/projects/www.worldcummunitygrid.org directory to limit the number of concurrent tasks for each project to the number of cores you actually have. If you have an 8 core machine and change ncpus to 16, the machine will run 16 tasks and it will over saturate your machine. The days calculation isn't exact so your machine may run dry before the bunker is due.
|
||
|
|
DCS1955
Veteran Cruncher USA Joined: May 24, 2016 Post Count: 668 Status: Offline Project Badges:
|
Looks like Team China does not Doneske as good as us. Took them two days to pass us up.
----------------------------------------4 Team China 15 years, 337 days, 17 hours, 55 minutes, 19 seconds 5 SNURK and friends 15 years, 255 days, 23 hours, 36 minutes, 53 Team China Avg. Run Time Per Calendar Day (y:d:h:m:s) 3:329:23:20:16 Ours Avg. Run Time Per Calendar Day (y:d:h:m:s) 1:339:10:17:48 Oh... they have 3331 members to our 50!!!!!!!!!!!!!!!!!!!!!  Based on the above.. Doneske Days Snurk and friends. 4661 days vs Team CHina 2964 days Basis Current Avg Day/day*2days Days Doneske Days SAF 1069 5730 4661 TC 2848 5812 2964 I know that avg day/day number is dated but it took TC an extra day to catch us up. [Edit 1 times, last edit by dcs1955 at Sep 25, 2019 3:04:08 AM] |
||
|
|
SNURK
Veteran Cruncher The Netherlands Joined: Nov 26, 2007 Post Count: 1217 Status: Offline Project Badges: 
|
I think it's safe to say that a top 10 spot in this challenge is secured
---------------------------------------- thanks again to a huge effort by all of you plus the special tactics described above.The upcoming third week of the THOR challenge (Monday October 07 to Sunday October 13) is going to be a real tough one. The target is top 3. Any ideas? |
||
|
|
DCS1955
Veteran Cruncher USA Joined: May 24, 2016 Post Count: 668 Status: Offline Project Badges:
|
Pray that MrKermit starts up again. + all of us use special tactics.
---------------------------------------- |
||
|
|
supdood
Senior Cruncher USA Joined: Aug 6, 2015 Post Count: 333 Status: Offline Project Badges:
|
Realistically, we should start discussions with the GPT. As we will both make it to the top 10, I think that we should both put in a good effort for the first half of the challenge, and then support whichever team looks like it has the best shot for top 3.
----------------------------------------Both teams are turning into "community teams" for the challenge (i.e., a lot of folks are leaving their usual teams and joining GPT or SNURK and friends for the challenge). I think that we should focus on getting one of the teams through to the top 3 and then all rally together to try to knock off one of the cryptocurrency teams. (And I'm fully aware that this will anger the conspiracy theorists who think that everyone is against the cryptocurrency teams.) |
||
|
|
PowerFactor
Ace Cruncher Joined: Dec 9, 2016 Post Count: 4030 Status: Offline Project Badges: 
|
You don't want to bunker now unless you plan on releasing them before Sunday. This week, just let them run and report. On Sunday after 00:00UTC, that's when you load up and turn off networking. Beware, you need to bunker the work that has a 10 day deadline(MIP ZIKA TB) or you may lose some work the following Sunday/Monday. The reason you are probably being limited to 5 days is because you are reaching the WCG imposed limit of 70 per thread. For an 8 core machine, that is 560 WUs (however long it takes your machine to run that). To get around that, you have to change ncpus in cc_config.xml to some higher value than your current number of cores. Here's the rub with that, you will also need to add (if you don't already have one) an app_config.xml file in your <BOINC DATA>/projects/www.worldcummunitygrid.org directory to limit the number of concurrent tasks for each project to the number of cores you actually have. If you have an 8 core machine and change ncpus to 16, the machine will run 16 tasks and it will over saturate your machine. The days calculation isn't exact so your machine may run dry before the bunker is due. I noticed that there is a 1000 WU limit for the boinc client which is independent of cpu threads. Doneske, have you overcome this limitation with app_config xml? |
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1324 Status: Offline Project Badges: 
|
I noticed that there is a 1000 WU limit for the boinc client which is independent of cpu threads. Doneske, have you overcome this limitation with app_config xml? This is a BOINC limitation. The only way to overcome is to run 2 or more BOINC-clients on 1 host. This can be a tricky process, causing a full loaded client to abandon all done tasks, when re-connect the server again. Be aware! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You don't want to bunker now unless you plan on releasing them before Sunday. This week, just let them run and report. On Sunday after 00:00UTC, that's when you load up and turn off networking. Beware, you need to bunker the work that has a 10 day deadline(MIP ZIKA TB) or you may lose some work the following Sunday/Monday. The reason you are probably being limited to 5 days is because you are reaching the WCG imposed limit of 70 per thread. For an 8 core machine, that is 560 WUs (however long it takes your machine to run that). To get around that, you have to change ncpus in cc_config.xml to some higher value than your current number of cores. Here's the rub with that, you will also need to add (if you don't already have one) an app_config.xml file in your <BOINC DATA>/projects/www.worldcummunitygrid.org directory to limit the number of concurrent tasks for each project to the number of cores you actually have. If you have an 8 core machine and change ncpus to 16, the machine will run 16 tasks and it will over saturate your machine. The days calculation isn't exact so your machine may run dry before the bunker is due. I noticed that there is a 1000 WU limit for the boinc client which is independent of cpu threads. Doneske, have you overcome this limitation with app_config xml? You can't overcome it (well, you could get the source and recompile the client) it is hard coded in the Boinc Client. I've been meaning to try it but haven't gotten around to it yet [Edit 1 times, last edit by Doneske at Sep 25, 2019 8:32:57 PM] |
||
|
|
|