Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
World Community Grid Forums
Category: Completed Research Forum: The Clean Energy Project - Phase 2 Forum Thread: CEP2 high checkpoint times |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 12
|
Author |
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Count your blessings with cep2 which is 'still' opt-in, the shuffling of jobs seemingly moving the very long run from job #2 to job #0. Running 5 qmc-cmn on a 4770. They have not checkpointed in 48 hours, progress percent going steadily up and then at some point resetting to 1.000 percent and sticking to that, though taskmanager gives them 99.9% utilization per thread. 4 suspended now to see how number 5 goes, at 39 percent and per boincstats still at checkpoint (0). If that one resets it becomes question time at their house of commons, speak forum. Not very inducing to compute for them and given there's no ups to bridge powerblips, 48 hours time 5 times 0 is nothing useful achieved. Given that the agent wont yield per normal switching routine long as there's no first checkpoint, it is most definitely an unbecoming elbow experience from qmc.
----------------------------------------[Edit 1 times, last edit by Former Member at Aug 9, 2014 5:03:11 PM] |
||
|
armstrdj
Former World Community Grid Tech Joined: Oct 21, 2004 Post Count: 695 Status: Offline Project Badges: |
The workunits do only checkpoint after each subjob completes. Unfortunately for larger molecules this can create very large time between checkpoints which is one of the reasons this project is opt in only and recommended to have leave application in memory set when running it. I believe for the new workunits they are a little front end loaded with larger molecules but we should start getting back into some smaller molecules soon.
Thanks, armstrdj |
||
|
|