Message boards :
Number crunching :
lunatics installation effects on other projects?
Message board moderation
Author | Message |
---|---|
merle van osdol Send message Joined: 23 Oct 02 Posts: 809 Credit: 1,980,117 RAC: 0 |
Does having lunatics 43 installed have an adverse effect when running Einstein, Milky Way or Collatz? merle - vote yes for freedom of speech |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
No, all the files added/changed by Lunatics installer are in SETI@home directory (<BOINC_Data>\projects\setiathome.berkeley.edu\) The only "adverse effect" may be if you run both apps at the same time and you set both SETI@home and Einstein to run on 0.5 GPU (2 tasks/GPU) In this case it may happen one SETI@home + one Einstein tasks to run on the GPU at the same time. Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
merle van osdol Send message Joined: 23 Oct 02 Posts: 809 Credit: 1,980,117 RAC: 0 |
No, all the files added/changed by Lunatics installer are in SETI@home directory (<BOINC_Data>\projects\setiathome.berkeley.edu\) Thanks BilBg merle - vote yes for freedom of speech |
David S Send message Joined: 4 Oct 99 Posts: 18352 Credit: 27,761,924 RAC: 12 |
No, all the files added/changed by Lunatics installer are in SETI@home directory (<BOINC_Data>\projects\setiathome.berkeley.edu\) I've had that happen occasionally. Does it hurt anything? Just Boinc allocating resources as it sees fit, right? David Sitting on my butt while others boldly go, Waiting for a message from a small furry creature from Alpha Centauri. |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
I've had that happen occasionally. Does it hurt anything? Just Boinc allocating resources as it sees fit, right? It does not "hurt", only affects Run times Unless driver crash or the GPU is too weak to be used simultaneously for 2 GPU apps from 2 different projects. If no errors happen it's safe. Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
David S Send message Joined: 4 Oct 99 Posts: 18352 Credit: 27,761,924 RAC: 12 |
I've had that happen occasionally. Does it hurt anything? Just Boinc allocating resources as it sees fit, right? Are you saying the GPU will do two different things less efficiently than two of the same thing? I've never had an error result from it, so I'm not (too) worried about that. David Sitting on my butt while others boldly go, Waiting for a message from a small furry creature from Alpha Centauri. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
I've had that happen occasionally. Does it hurt anything? Just Boinc allocating resources as it sees fit, right? In the early years of BOINC, using CPUs only, we sometimes found that running different applications on different cores of a Duo or Quad CPU worked to the benefit of both projects - presumably the two applications placed different demands on the shared components of the CPU, and kept out of each others way. With multiple compute resources in the same computer nowadays (and with iGPUs, even on the same silicon), it's harder to predict what will happen. You'll have seen me working on the AP v7 work fetch issue recently, using both facilities of my i5 Haswell CPU (host 7118033). Usually, that host runs other CPU-only projects on its CPU cores, and either SETI or Einstein on the iGPU. For the AP test, I replaced the other projects (NumberFields and SIMAP) on the CPU with the highly-optimised 64-bit AVX Lunatics AP application. I noticed that the iGPU slowed down significantly when I did that - AP task runtimes increased by some 30%, from ~10 Ksec to ~13 Ksec. I'm going to switch that back when the current CPU cache has drained, and verify that it's a genuine reversible effect. So, even among CPU applications, there are what I call 'heavy' and 'light' programs, and choosing between them can significantly affect other parts of the system. I imagine the effect is most pronounced on modern CPUs with sophisticated internal power and thermal management tools embedded - core clock speeds can be changed dynamically to keep within the TDP envelop, and that can affect throughput noticably. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
I noticed that the iGPU slowed down significantly when I did that - AP task runtimes increased by some 30%, from ~10 Ksec to ~13 Ksec. I'm going to switch that back when the current CPU cache has drained, and verify that it's a genuine reversible effect. And discussion regarding this particular effect is here: http://setiathome.berkeley.edu/forum_thread.php?id=75971 |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
That other discussion is on the opposite effect, "that using the iGPU on Haswell ... slowed CPU processing significantly" Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
That other discussion is on the opposite effect, I think they are interrelated. Big load of one APU part slow downs the other. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.