Questions and Answers : Windows : Does dirty reboot spoil Rosetta results?
Author | Message |
---|---|
Denys Romaniuk Send message Joined: 2 Dec 10 Posts: 4 Credit: 335,885 RAC: 0 |
Hello, world! My Windows10 computer sometimes shows Blue Screen and reboot for unknown reason. I don't know how to fix this. Question: is it ok to run Boinc and Rosetta@home on this computer? Can the dirty reboot spoil running tasks of Rosetta? Thanks for your replies. |
Grant (SSSF) Send message Joined: 28 Mar 20 Posts: 1681 Credit: 17,854,150 RAC: 22,647 |
My Windows10 computer sometimes shows Blue Screen and reboot for unknown reason.I'd suggest checking the CPU temperature when under load. If it's OK, then running a memory Test would be the next step. If the memory's Ok then the next step would be to check the power supply rails and make sure the voltages are correct. Crashes that take a while to occur, CPU temperature & power supply are most likely. Very random crashes, memory or power supply are most likely. Really, really random crashes, memory or a driver issue. Question: is it ok to run Boinc and Rosetta@home on this computer?Very possibly. Same as for any programme that is running when a system crashes. Grant Darwin NT |
Denys Romaniuk Send message Joined: 2 Dec 10 Posts: 4 Credit: 335,885 RAC: 0 |
Thanks for your reply, but. one more question... On the Rosetta site, in Results section.... 99% of my tasks are shown as "Completed and checked". Does "checked" mean that the result is totally and 100% correct? |
Mod.Sense Volunteer moderator Send message Joined: 22 Aug 06 Posts: 4018 Credit: 0 RAC: 0 |
Typically tasks take a "checkpoint" every 15 minutes or so (not always, not typically). They ALSO take a checkpoint at the end of a model. When a task is shutdown for any reason and forced to restart, this restart occurs from the last checkpoint. So, your odds of actually corrupting one of the results is pretty slim. And even if you did, it is not a big deal. The whole scheme here is to compute hundreds of thousands of models, look for the best results and then reconfirm all of those best results. So if you happen to corrupt things in a way that looks like a bad results, it just goes in to the pile of things that don't look promising. If you happen to corrupt things in a way that looks like a fantastic result, that specific model is rerun and reconfirmed and easily found to be flawed. But, I would suggest you open a thread on the Number Crunching board and ask for help with your BSD (blue screen of death) problem. Rosetta Moderator: Mod.Sense |
Questions and Answers :
Windows :
Does dirty reboot spoil Rosetta results?
©2024 University of Washington
https://www.bakerlab.org