Posts by Mike Gelvin

1) Message boards : Number crunching : Problems with minirosetta version 1.+ (Message 51452) Posted 17 Feb 2008 by Mike Gelvin Post: Sadly I have to stop running Rosetta. I am getting way too much grief from MiniRosetta vs NOD32. Ill keep running Ralph on one system and wait until this gets sorted out, and then I shall return. I'm disappointed in the response that has allowed this application to migrate and/or continue on Rosetta when this issue was identified on Ralph and not addressed. Reminds me of Predictor@Home.
2) Message boards : Number crunching : Problems with minirosetta version 1.+ (Message 51275) Posted 9 Feb 2008 by Mike Gelvin Post: NOD 32 Antivirus is blocking the download. It says it is a virus? I got the same problem on Ralph. 1.05 and 1.07 were identified as being a probable virus.
3) Message boards : Number crunching : Credits/CPU Second (Message 42220) Posted 16 Jun 2007 by Mike Gelvin Post: I would still like this metric to work. Any word?
4) Message boards : Number crunching : Problems with Rosetta version 5.67 (Message 41511) Posted 26 May 2007 by Mike Gelvin Post: I have a work unit that seems to be "almost" stuck at 97% complete. The % complete has been slowly increasing (by about .4%) over the last two hours. I have work units set to complete in 4 hours, and we are going over 6 with this one. It is wuid=74278854
5) Message boards : Number crunching : Problems with Rosetta version 5.62 (Message 41435) Posted 25 May 2007 by Mike Gelvin Post: Do work units with large memory requirements get granted more credit/hr due to the increase in resources consumed? I can sure tell when one of them is on my system (or worse, two of them at the same time). I understand I could limit the amount of memory consumed, but would rather not.
6) Message boards : Number crunching : Credits/CPU Second (Message 41401) Posted 24 May 2007 by Mike Gelvin Post: I believe this metric has not been update since before February. Is there a reason for not keeping this stat up to date? I think some of the boards are erroneously using this information.
7) Message boards : Number crunching : Credits/CPU Second (Message 41304) Posted 22 May 2007 by Mike Gelvin Post: Looks like Mike is referring to a file on the server. Here's some documentation on the subject I take it you know how to interpret the time format listed there, which shows <update_time>1179785967</update_time> I've sent an EMail to David Kim requesting he or another member of the team review how/when these statistics get updated. I believe this file gets update every day. It is an entry within the file that I believe does not get update. That entry is named <credit_per_cpu_sec>.
8) Message boards : Number crunching : Rosetta v Seti times v credit (Message 41284) Posted 22 May 2007 by Mike Gelvin Post: According to the stats site, Seti provides 1.084 times the credit Rosetta provides, comparing 22000 computers that are crunching both. I do not think this is accurate. I believe the web page you got the figure from is based on information contained in the Hosts.gz file provided by Rosetta. In that file, the specific information (<credit_per_cpu_sec>) has not been update by Rosetta for a LONG time, hence the stat is probably wrong. I’m actively seeking an answer to this in another tread.
9) Message boards : Number crunching : Credits/CPU Second (Message 41271) Posted 21 May 2007 by Mike Gelvin Post: Anyone have the answer to the question in this thread?
10) Message boards : Number crunching : Credits/CPU Second (Message 41111) Posted 17 May 2007 by Mike Gelvin Post: Are you talking about a statistic that is in your host's details page? Or are you talking about the chart on this website: http://boinc.netsoft-online.com/e107_plugins/boinc/get_cpcs.php? I'm talking about the xml element like " <credit_per_cpu_sec>0.001260</credit_per_cpu_sec> " found in the host.gz file. Which i believe does power the web page you were refering to.
11) Message boards : Number crunching : Credits/CPU Second (Message 41077) Posted 16 May 2007 by Mike Gelvin Post: I ask, because the numbers have not changed at all in 2 weeks. Does anyone know? Is this broke?
12) Message boards : Number crunching : Credits/CPU Second (Message 40549) Posted 8 May 2007 by Mike Gelvin Post: How often does the Credits/CPU Second stat get update in the hosts file?
13) Message boards : Number crunching : Problems with Rosetta version 5.62 (Message 40478) Posted 7 May 2007 by Mike Gelvin Post: I have noticed several work units now take 450Meg + RAM during execution. Has the system requirements changed? Specifically the "FRA_a016_STRUCTURAL_GENOMICS_hom001_2_a016_2_1i74A_IGNORE_THE_REST" series.
14) Message boards : Number crunching : Problems with Rosetta version 5.40 (Message 31870) Posted 30 Nov 2006 by Mike Gelvin Post: This result did not progress beyond 20 seconds on a remote computer for 9 days DOC_1FSS_R061114_pose_u_global_search_1402_1511 http://boinc.bakerlab.org/rosetta/workunit.php?wuid=42483747 I stopped BOINC, and restarted, it froze again at 19 seconds I aborted the work unit. ... Mike, the computer you reference has only 128 MB of RAM. I have a small laptop crunching Roseta that has only 168 MB, and for some work units, it can take a very long time to show progress. Also, you might want to check the amount of cache and increase it if possible. I try to run around 1 GB on the laptop. It's slow, but persistent! :-) I'm not talking about progress... I'm talking about CPU time accumulated. It froze at 20 seconds. Edit: Thanks for pointing out the 128 issue. I have detached this computer from the project.
15) Message boards : Number crunching : Problems with Rosetta version 5.40 (Message 31826) Posted 29 Nov 2006 by Mike Gelvin Post: This result did not progress beyond 20 seconds on a remote computer for 9 days DOC_1FSS_R061114_pose_u_global_search_1402_1511 http://boinc.bakerlab.org/rosetta/workunit.php?wuid=42483747 I stopped BOINC, and restarted, it froze again at 19 seconds I aborted the work unit. Work units that hog the resources of a computer are UNACCEPTABLE! This is the second time in a few months that Rosetta has messed up on the production (not Ralph) site. One more time and I shall pull my support of Rosetta. Edit: I think I have another system out there that has been hung for 29 days but since you seem to purge results going back that far, I cant tell if its Rosetta that's the problem. I am not scheduled to visit that site for another 3 weeks, so it looks like that system is useless to ALL BOINC projects until I can clear you out. A scary thought is that some of my remote computers are hyper-threaded, or have multi-processors. A Rosetta task could take down one of the cores and yet the computer will still report into projects and I won’t be able to tell.
16) Message boards : Number crunching : Report problems with Rosetta version 5.36 (Message 30837) Posted 9 Nov 2006 by Mike Gelvin Post: I have a 5.36 work unit on a remote system that appears to have hung for 8 days with no additional CPU time past the first 2 hours 31 minutes. This system does not run a screensaver, and has BOINC installed as a service. http://boinc.bakerlab.org/rosetta/result.php?resultid=44919383 Why hasnt the watchdog killed this work unit?
17) Message boards : Number crunching : Report problems with Rosetta version 5.32 (Message 29958) Posted 24 Oct 2006 by Mike Gelvin Post: running 5.32 Workunit: http://boinc.bakerlab.org/rosetta/workunit.php?wuid=38429684 It has frozen at 46 minutes of runtime, I suspect this occured 27 hours ago as my system idle time is up to 27 hours now. I have shut BOINC down and restarted. the same work unit is now running again from 0. Result errored out after about 12 minutes. Dump is avaliable in result page. An interesting thing to note, is that when it "hung" it dumped with an error of LoadLibraryA(srcsrv.dll): GetLastError = 126 followed by an access violation. Then it hung. So, there are 2 dumps in the file. I would have thought they would be a bit more agressive on Ralph taking care of the problems posted in this thread. I dont like these kinds of problems on my production machines, but I do have a machine that Ralph runs on to help ferret out these problems.
18) Message boards : Number crunching : Report problems with Rosetta version 5.32 (Message 29953) Posted 24 Oct 2006 by Mike Gelvin Post: running 5.32 Workunit: http://boinc.bakerlab.org/rosetta/workunit.php?wuid=38429684 It has frozen at 46 minutes of runtime, I suspect this occured 27 hours ago as my system idle time is up to 27 hours now. I have shut BOINC down and restarted. the same work unit is now running again from 0.
19) Message boards : Number crunching : What makes a mobo "server" quality ? (Message 26732) Posted 14 Sep 2006 by Mike Gelvin Post: Usually you get more than one ethernet jack, and the ability to net boot. Check out the bios section of the manual, and you will see some of the unique features. Also server boards usually come with built in RAID, and some kind of independent monitor processor, so you can monitor/reboot remotely a board that is totally hung.
20) Message boards : Rosetta@home Science : CASP refinement (Message 22458) Posted 14 Aug 2006 by Mike Gelvin Post: See "David Baker's Rosetta@home journal" thread on this board for later info regarding this. From what I gather from other posts, the refinement phase will run until around the 28th (I am not sure what the exact date is).

Next 20