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: 3513
|
![]() |
Author |
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12561 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Unixchick
----------------------------------------My spreadsheet covers all you ask for. There are 35,609 squares all told. 318 are stuck in extremes (0.9%), 441 stuck in accelerated (1.2%) and 1,180 in generations 137 to 141, most of which are stuck (3.3%). However, we have 1,433,434 units remaining to crunch of which 18,044 (1.3%) are extremes, 21,709 (1.5%) are accelerated and 51,467 (3.6%) are in generations 137 to 141. The second set of figures are higher because they have further to go. All figures at 12:00 GMT (UTC) and the remaining units allow up to generation 182 to complete the year. I currently have 11 units of which 8 are crunching (2 machines). Mike [Edit 2 times, last edit by Mike.Gibson at Apr 14, 2025 11:35:36 PM] |
||
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 1114 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() |
Thanks Mike.
That helps me understand how much of the data is lost vs kept if they decide to abandon the ones not moving (I hope they don't abandon them all, but helps me to understand the impact on science if they do). Of course we will definitely find more stuck ones as we move through the generations. |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12561 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Unixchick
The number suspected as being stuck in generations 137 to 141 might be because they haven't taken the first stage of reducing the TimeStep. That also might be the problem from generation 125 onwards. There are only 23 stuck units before that generation. Mike |
||
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 1114 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() |
I certainly hope not all the stuck ones are truly stuck, but I wanted to know the worst case scenario.I was also trying to figure out if taking the stuck ones out of the calculation would change the finish date. As the end date has been a discussion in this thread lately.
I know the IT team is busy keeping WCG functioning on old hardware and working on MAM, but I really do hope we can have an official ARP update soon. I am grateful we have a steady flow of ARP WUs now though. I really do appreciate the time you take to look at the info and give us weekly and daily updates. Thanks, Mike ! |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12561 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Unixchick
You were referring earlier to discrepancies in the numbers in state.txt, I presume. It might well be because of units getting stuck in, say, normals and for that generation having subsequently been classified as accelerated or extreme. There are 35,161 listed so 448 are missing. This is similar to the number stuck in accelerated, but could be the total relates to all the moverssince a certain event. My stats ignore state.txt. Mike |
||
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1311 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I currently have 11 units of which 8 are crunching (2 machines). Mike when you are able to if the others 3 units are not processing are you able to manually start them/move them to the front of the queue? Thanks ![]() |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12561 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Speedy51
I could but I won't. Both machines are i7 3770. They are each set for a cache of 12 ARP units and to process 4 at a time. This conforms to 50% of threads. The other 4 threads on each machine crunch MCM. I want to crunch both projects and if I crunch more ARP they slow down. Each ARP unit takes about 1 day so if I happened to get 12 on each machine (very rarely), they would all finish within 3 days which is half the deadline. Mike |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12561 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Again no movement of extremes or accelerated. There are 759, all of which appear to be stuck.
We are now 29% into generation 143. 1,360 normals validated yesterday from 142 & 143 only. There are now 29,323 normals in the generations being released, many of which may be stuck, particularly in generations 137 to 141. There is at least 1 moving in generations 137 & 141. There are now 4,809 units held up in generation 144. Mike |
||
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1311 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
phytell I would suggest that if you have more than 2 threads on that machine that you suspend the 2 newest running MCMs and those ready to start before your ARPs. That would start them. You could then resume the suspended units. You would then have the 2 ARPs running together with some MCMs. Mike Mike I was just replicating the above post on page 160. (Asking you to start any ARP work that had not been started) I figure what is good for the goose is good for the gander :-) ![]() |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12561 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Speedy51
The circumstances were not as you described. I prioritise ARP to a maximum of half threads so as not to slow down crunching. phytell was trying to do the same for 2. Mike |
||
|
|
![]() |