Posts by E01902

1) Message boards : Number crunching : Machines runs out of units every few minutes (Message 63032) Posted 17 Jan 2015 by E01902 Post: Actually it's important that the project leader(s) hear that people are frustrated and trying to cheat the system. This allows them to manage the cheating. Someones telling them they're calling in extra because xxx. Now they now why there's an extra load on that server without having to research for hours. If people can't get enough work they'll move on to other projects. So, either this project has more processing power than it needs, which is a good position to be in, or it has a project problem where it can't turn work around fast enough and the developers need to explain what's going on. Communication from the Project Leader is just as important as developing the code to fix the problem. Thanks for reminding people of the impact of what they think might be a quick fix, doesn't necessarily help the system if everyone tries the same thing.
2) Message boards : News : New Nbody version 1.46 (Message 62995) Posted 12 Jan 2015 by E01902 Post: The time Remaining estimate went from 18 to 131 hrs overnight. This machine would never give enough processing time by the task deadline, so I aborted it.
3) Message boards : News : New Nbody version 1.46 (Message 62994) Posted 12 Jan 2015 by E01902 Post: Does this mean the tasks are going to really take longer than the Remaining estimate? I have a task running all day that is now 9% complete. It's recorded 14 hr run time and somehow it only thinks it has 18 hours left. The other thing that's odd is, I don't think that 18 hours has counted down much during the day but I didn't write down the initial number and don't do this enough to know where to look. Task manager thinks it's working on something, so hopefully its not running in circles. I'll let it run over night and check on it in the morning.
4) Message boards : Number crunching : Milkyway ignoring Boinc CPU limit (Message 62992) Posted 11 Jan 2015 by E01902 Post: The Einstein project tasks that I've run never use more than one core each and I've noticed they interpret a 50% CPU limit as 50% per core. So if you suspend two tasks the other two don't ramp their two cores to 100%ish to put the average load at 50%, which I always thought was their loss. A single four core task on MilkyWay is limiting to 50% avg but it seems to ignore the number of cores limit. Either I shouldn't have gotten that task or it should have only run on two. I'm not sure how the tasks are written. I haven't been running MW long, but have had a couple of single core tasks come through. I just haven't observed their interpretation of the rules. I was hoping to say here's your two cores Milkyway, and here's your two cores Einstein, but it seems I'll just have to let them run to their own interpretation and see if the task switch time works. What started all this was the four core MW task jumped right in and took over the four cores and set a couple of Einsteins tasks to waiting to run or some verbage. This may be because BOINC manager would have long since seen the task switch timer expire since Einstein was the only thing running. Unless I suspended the MW task there was no way to get it to share. I'll check on it later, if MW doesn't share with the task time switch set, it'll have to go.
5) Message boards : Number crunching : Milkyway ignoring Boinc CPU limit (Message 62981) Posted 10 Jan 2015 by E01902 Post: I have a 4 core cpu which I'm not sure if BOINC considers that 1 CPU or 4. Anyhow I set a limit of 50 % of the CPUs and 2 CPUs which I thought would limit Milkway to 2 cores but clearly one of the tasks says 4 cores and task manager shows its working with all 4. So is Milkyway ignoring the Boinc rules, or is there something I don't understand? Both are likely.