Posts by old_user58799

1) Message boards : Number crunching : HADSM3 slab 6.07 model exited zero status & no finished file (Message 35391) Posted 29 Oct 2008 by old_user58799 Post: I\'m only half way through, 500 hours, should I continue? I\'m getting these either 15 minutes apart, or several hours.
2) Questions and Answers : Wish list : Running multiple climate models on dual processor CPUs (Message 33866) Posted 19 May 2008 by old_user58799 Post: True, this is a Hyperthreading CPU, really, I wasn\'t aware it was only 1 CPU. When I run only a single task, Windows XP task manger says that is only getting 50% of my machine, so I needed to tell BONIC to use 2 processors in it\'s options to be able to get to 100%. What is the CPDN screensaver? The Windows XP screensaver is set to Blank, and kicks in after 3 minutes. At one time, they had recommended not to leave the BONIC Manager in the Tasks or Download panel, but rather switch to either Projects or Messages, to save the CPU cycles when refreshing those %\'s, even in if the application is minimized, is that still true? Runtime debt? Oh yeh, read a little about that a long time ago, do I need to be concerned? Guess I\'m afraid if I detach from a project, might not get reconnected again, as I\'ve used different email address over the years for different projects, some which might no longer be valid. The earlier post about the countdown number on the Show Graphics screen, that\'s neet. How would one ever guess that \"Z\" and \"8\" would do that, what else will is do, is it in some document somewheres? A \"Time to next checkpoint\" or \"Date/time of next checkpoint\" would be nice. How about a \"Suspend After Checkpoint\", so BONIC stops the process at the optimum point, for when we need to restart BONIC or reboot the computer?
3) Questions and Answers : Wish list : Running multiple climate models on dual processor CPUs (Message 33762) Posted 14 May 2008 by old_user58799 Post: I have a dual processor computer, 2.8 GHz. When I get a \"C\" model, while it says it will take 4000 hours to complete, but it took 800 hours to get to 10% complete. Therefore, I might assume it will take 8000 hours to complete this model. To help achieve that, I\'d like to dedicate one processor to just running the \"C\" model. The other processor I\'d like to rotate between a climate \"S\" model, and some of the other dozen BONIC applications. I can fake this, in a clumsy manner, by suspending all but 2 models on my computer, the \"C\" model, and what ever other application I want to run. But I need to time it, so I\'m home when the computation ends, so I can Resume the next application. If I\'m wrong, a CPU goes un-used for a period of time. If I leave Network Activity Always Enabled, then I also need to change the Status of each project to Won\'t Get New Tasks to prevent new tasks being downloaded and taking the processor away from the \"C\" model.
4) Questions and Answers : Wish list : Trickle up messages don\'t identify which process produced it (Message 33761) Posted 14 May 2008 by old_user58799 Post: I\'m on a dual CPU processor, and running 2 climate models, one \"C\" and one \"S\". But when I see trick up messages in the log, I can\'t tell which process produced it.
5) Message boards : climateprediction.net Science : Disk space consumed & exceeded with Sulpher model (Message 20117) Posted 10 Feb 2006 by old_user58799 Post: Humm, this could be interesting. How does one \"look at the page for that model\"?
6) Message boards : climateprediction.net Science : Disk space consumed & exceeded with Sulpher model (Message 20113) Posted 10 Feb 2006 by old_user58799 Post: Thanks. 1 GB is correct. Rather than using my rotating drive, I\'m using either a USB memory stick memory, or a compact flash type of solid state drive, and 1 GB is the max size of what I have (and I\'m allowing BOINC to use it all). I have no idea if Phase I was completed within that first 10 to 12% before the 1 GB was consumed & abended. Seems like I should temporarly suspend (or not accept new work) myself from the climate project until those new models are available. Charlie
7) Message boards : climateprediction.net Science : Disk space consumed & exceeded with Sulpher model (Message 20106) Posted 10 Feb 2006 by old_user58799 Post: How much disk space does the Sulpher model use, and what happens if that exceeds the amount of disk space I have? I have a 1 GB hard drive, and being generous, I allow BOINC to use most of it. However, I have noticed that when the Sulpher model gets about 10% to 12% completed, it has used about 1 GB, and my model fails. It has done this now for 3 models. Doesn\'t BOINC check to see if the user has enough disk space to hold the expected amount of output? Is there any value in having this partly completed models, which I presume now someone else is going to have download and recalculate what I have done and take them to completion? Is there some way to not download the Sulpher models, and only download the smaller models?
8) Questions and Answers : Windows : Bonic consumes too much CPU, boggs down other apps; or doesn\'t get any CPU time (Message 16977) Posted 4 Nov 2005 by old_user58799 Post: I have used classic SETI for many years, and setting the Client Priority to LOW, I\'m able to work on other apps, and not realize it is taking spare cycles on a moment to moment basis (as reported by the Task Manager window), so the System Idle process gets very little. However, with Boinc (4.45 & 4.19), if that is set to run all the time, I notice that it does bogg my PC down considerably, such that I need to manually suspend all projects while I\'m trying to use my computer, especially for the occassionally CPU intensive apps. The preferences on this computer (in Boinc) is probably to run all the time, and to remove from memory. A somewhat separate issue: Usually during a day at work, I can complete 3 or 4 classic SETI WU\'s. Yesterday, running BOINC with Climate & Protein apps., no work was completed at the end of an 9 hour day, with light usage of the PC. The Boinc preferences on this is set to wait 3 minutes after last use of computer, and to remove from memory. I\'m wondering if the \'remove from memory\' causes temperary & intermediate results to be lost, so the next time it comes in to memory, it has to begin at the beginning of the last \'checkpoint\'. Perhaps the remain in memory, just be paged out, is much better for this? Thanks Charlie