1)
Message boards :
Number crunching :
New series of invalids
(Message 68996)
Posted 31 Aug 2019 by TB Post: My CPU is slow. When I see a batch that has fairly high levels of invalid responses, as soon as they download I abort them and chuck them back into the "available for sending" pool. That's about it. It's a CPU thing, which means ultimately it's a software issue, either in the code or depending on some difference in how the different CPUs crunch numbers. |
2)
Message boards :
Number crunching :
NoContraintsWithDisk200 vs. south4s (or whatever)
(Message 68534)
Posted 15 Apr 2019 by TB Post: They haven't crashed for me. NoConstraintsWithDisk200 have worked fine--hadn't noticed the change up. Noticed "south" just now (first one's I've seen) so I searched after the mess last month (or was that February?) and saw your post. Trying to run them now. No explanation, but thought you'd like to know it's not a general problem. |
3)
Message boards :
Number crunching :
Very long N-Body simulation
(Message 67854)
Posted 30 Oct 2018 by TB Post: Dunno. I've gone back and said I don't want to run n-body simulations after this batch. At least for a while--I have to assume that "they" didn't plan for work units to be this large. |
4)
Message boards :
Number crunching :
Very long N-Body simulation
(Message 67852)
Posted 29 Oct 2018 by TB Post: I've had some like that. Some just took 5-9 hours. One went 17 hours. Currently there's this thing that has crunched for 10 hours and claims to have 17 to go: Application MilkyWay@Home N-Body Simulation 1.72 (mt) Name de_nbody_10_24_2018_v172_20k__data_1_1540379102_15905 State Running Received 10/27/2018 10:57:06 AM Report deadline 11/8/2018 9:57:05 AM Resources 4 CPUs Estimated computation size 86,339 GFLOPs CPU time 1d 15:23:52 CPU time since checkpoint 00:05:19 Elapsed time 10:39:53 Estimated time remaining 17:07:32 Fraction done 38.376% Virtual memory size 40.31 MB Working set size 37.50 MB Directory slots/4 Process ID 24988 Progress rate 3.600% per hour Executable milkyway_nbody_1.72_windows_x86_64__mt.exe Don't like them. If there's an error it means a big chunk of time's devoted to something that's corrupted. It means that somebody else has to take another big chunk of time to crunch the numbers a 3rd time. |
©2024 Astroinformatics Group