Welcome to MilkyWay@home

Posts by Aurum

1) Message boards : Number crunching : CPU versus GPU (Message 68972)
Posted 13 Aug 2019 by Aurum
Post:
This is what I want:
✔ Milkyway@home N-Body Simulation - CPU
Milkyway@home Separation - CPU
✔Milkyway@home Separation - GPU
2) Message boards : Number crunching : CPU versus GPU (Message 68970)
Posted 12 Aug 2019 by Aurum
Post:
I just noticed that N-Body Simulation is only CPU. So I checked the box Use CPU. But then I get both CPU & GPU WUs for Separation. Is there any reason to run both CPU & GPU Separation WUs???
Could another option be added to Preferences/Run only the selected applications:
✔ Milkyway@home N-Body Simulation
Milkyway@home Separation - CPU
✔Milkyway@home Separation - GPU
3) Message boards : Number crunching : MilkyWay takes a backseat to Einstein ??? (Message 68968)
Posted 11 Aug 2019 by Aurum
Post:
You will find running the same project on your cpu and gpu will cause problems too as they do not have separate caches.
I did not know that. What cache is that?

I was hoping Dr Jake would just adjust a server setting and put them on a level playing field. But as soon as I posted I realized I'd be told to adjust Resource Share. I never use it since it can have undesired results and besides with multiple computers I'd lose track of which was set to what.

If you watch GPU Load using GPU-z you see that MilkyWay is a short duty cycle while Einstein is medium and GPUGrid is long. By pairing longer projects with MilkyWay it gets a more efficient and complete use of the GPU.
I recently saw a post where someone was using time stamps in the logs to measure the time over multiple WUs for Linux computers to test if pairings produced more points per hour.
4) Message boards : Number crunching : MilkyWay takes a backseat to Einstein ??? (Message 68963)
Posted 11 Aug 2019 by Aurum
Post:
I like to run MilkyWay with Einstein WUs on the same GPU, a perfect pairing. However, Einstein will continue to request work but MilkyWay stops requesting work and runs out. To get more MilkyWay WUs I have to Suspend Einstein to stock up on MilkyWay.
Is it possible to make MilkyWay as aggressive as Einstein???
5) Message boards : News : 30 Workunit Limit Per Request - Fix Implemented (Message 68491)
Posted 8 Apr 2019 by Aurum
Post:
I never get enough MW WUs. I can run 200 at a time. Some computers don't get any while some others get a steady supply. They all have the same app_config.xml. Server Status shows 10,000 WUs ready but I have computers that haven't gotten any in forever.
Is there some kind of governor on this project???
I can't find anything to explain the difference.
6) Message boards : Number crunching : Linux Mint 19 Tara (Message 67676)
Posted 18 Jul 2018 by Aurum
Post:
I believe I fixed it.
First: Linux 19 does not default to include recommended dependencies. Fix with 1.2.1: https://sites.google.com/site/easylinuxtipsproject/mint-cinnamon-first
Second: Install CUDA: https://help.ubuntu.com/community/Cuda
7) Message boards : Number crunching : Linux Mint 19 Tara (Message 67674)
Posted 17 Jul 2018 by Aurum
Post:
I upgraded a computer from Linux Mint 18.3 Sylvia to Linux Mint 19 Tara and now BOINC will not run GPU WUs any more. I'm running BOINC 7.12.0. The only Notice is:
Milkyway@Home: Notice from BOINC
Your current settings do not allow tasks from this project. To fix this, you can change Project Preferences on the project's web site.
Tue 17 Jul 2018 12:14:47 AM PDT
Anyone know how I can get GPUs working again???
BTW, it took several hours to upgrade to Linux 19. NoMachine works so much faster on this headless computer than Linux 18.3 which was unbearably slow.
I don't recommend upgrading until more kinks get worked out.
Tried Folding@Home 7.5.1 & it DLs WUs but shortly after they start running they fail and another WU is DLed. All 3 1080 Ti's do the same. I have all the latest updates and the latest Nvidia driver.
8) Message boards : Number crunching : Frequency preference (Message 67673)
Posted 17 Jul 2018 by Aurum
Post:
Grammar seems ambiguous, could be worded better. Seems counter intuitive. Shouldn't higher frame rates run WUs faster???
I'm trying 30 fps but I don't see any difference. Does this parameter work???
Frequency (in Hz) that should try to complete individual work chunks. Higher numbers may run slower but will provide a more responsive system. Lower may be faster but more laggy. 
default 60 (corresponds to 60 fps)
9) Message boards : News : New nbody runs July 2 (Message 67650)
Posted 3 Jul 2018 by Aurum
Post:
I never get any nbody WUs. I have both checked in my milkyway preferences. Any idea why not???




©2019 Astroinformatics Group