AK V8 + CUDA MB team work mod

[Mach]

Ok, after a little more monitoring, I've come to the conclusion that when the above mentioned problem shows up (WU utilizing the GPU is swapped out for another WU of higher priority, leaving the GPU in an idle state), the GPU is never reloaded with another WU.  I've monitored the process and even after all three of the WUs complete, more WUs are started, but never moved to GPU processing.  It's almost as if the GPU is never flagged as free when the WU is swapped out and the team mod is sending any further work to the CPU cores.

[Raistmer]

That mod sounds exactly what I need in this case.  Unfortunately, my CPU only supports the SSE3 instruction set and not SSSE3    Oh well, guess I'll just have to live with the occasional idle GPU for the time being.  Thanks for the prompt responses above, Raistmer 

Don't worry, soon will be SSE3 support added too, just check this thread time to time.

[Raistmer]

It's almost as if the GPU is never flagged as free when the WU is swapped out and the team mod is sending any further work to the CPU cores.

You can check this: try to open file gpu_file_lock in SETI@home project directory when your GPU is idle.

[Geek@Play]

Raistmer............
Running this "team" version of software.  Yesterday found one of my quad computers running 5 Astropulse work units on the cpu's.  I paused one of the AstroPulse work units and resumed it later in order to start up a cuda work unit.  Has this been reported before?

[Raistmer]

Raistmer............
Running this "team" version of software.  Yesterday found one of my quad computers running 5 Astropulse work units on the cpu's.  I paused one of the AstroPulse work units and resumed it later in order to start up a cuda work unit.  Has this been reported before?

Yes. It can't be repaired w/o direct interaction with BOINC. So as I said many times, kick BOINC devs for native CPU + GPU support for same app. They do this for srock apps already, now they should provide the way to do the same with app_info file installed.

[Marius]

Playing games like WOW and doing cuda in the background doesn't go very well. Is there a way to disable/suspend only cuda while playing?

something like a semaphore file?

No
Try to kick BOINC devs some more for it

That must be you as i'm running apps from you for over 1 year now
solution: c:\ut\pssuspend.exe MB_6.08_mod_CPU_team_CUDA.exe

[Raistmer]

Playing games like WOW and doing cuda in the background doesn't go very well. Is there a way to disable/suspend only cuda while playing?

something like a semaphore file?

No
Try to kick BOINC devs some more for it

That must be you as i'm running apps from you for over 1 year now
solution: c:\ut\pssuspend.exe MB_6.08_mod_CPU_team_CUDA.exe

Sure you can suspend process via ProcessLasso, via ProcessExplorer or by any other means... but it's OS level solution, not BOINC level one.

[Barnie3]

I have a small problem, may be noobish, don't know.
My client doesnt request new WUs. It always ask for 0 seconds of new work although the cue is nearly empty now and it should be enough for 8 days (tried different settings here)
A few days ago i installed the third package of first post of this thread on my intel dualcore. Once in a while i had AP WUs in my cue so i edited the app_info.xml and cutted the lines referencing AP. It worked nicely thereafter, but now ... no more WUs at all. Certainly i am not sure if there is a connection with my editing as it does not help to restore the original app_info.xml as provided by raistmer. Any ideas ?

p.s. sry for my bad english

edit:

i had to learn that the client does not request new tasks if there are too much unsent results waiting to be transferred. In the net i've read that too much is something like more than two times the number of your cores. I can't proove this, but from the behavior of my clients this seems to be reasonable.

[andreas]

Hi all,

I'm running the AK V8 + CUDA MB team mod since last week, and everything was fine till the weekend. The WU's from my cache worked fine, but after finaly getting new work, I got quite a lot compute errors withs BOINC telling me "Outpute file xy_0 for task xy is absent". On one machine I get these errors for about 1 out of 20 WU's, but the other is much worse (so I switched it to get no new work).

Any suggestions for this problem?

BTW, when I started the mod the first time I observed the following:
The 6.03 WU's from my cache were processed fine, but all new work was 6.08 WU's (prcessed fine as well). As I understand it, the team mode should be able to process both types (as the are not realy different, just another name basically), then why is my BOINC just getting 6.08 WU's and not mixed work, depending on availability? This is not a problem, I'm just curious.

And Raistmer and all involved, keep up the good work. It realy is appreciated.

Andreas

[Raistmer]

Hi all,

I'm running the AK V8 + CUDA MB team mod since last week, and everything was fine till the weekend. The WU's from my cache worked fine, but after finaly getting new work, I got quite a lot compute errors withs BOINC telling me "Outpute file xy_0 for task xy is absent". On one machine I get these errors for about 1 out of 20 WU's, but the other is much worse (so I switched it to get no new work).

Any suggestions for this problem?

BTW, when I started the mod the first time I observed the following:
The 6.03 WU's from my cache were processed fine, but all new work was 6.08 WU's (prcessed fine as well). As I understand it, the team mode should be able to process both types (as the are not realy different, just another name basically), then why is my BOINC just getting 6.08 WU's and not mixed work, depending on availability? This is not a problem, I'm just curious.

And Raistmer and all involved, keep up the good work. It realy is appreciated.

Andreas

1) Probably your BOINC installation was unable to load files from berkeley's servers. Network connection overloaded now so corrupted downloads is very likely.
2) BOINC will mark all new work with latest app version available to it.

[andreas]

1) Probably your BOINC installation was unable to load files from berkeley's servers. Network connection overloaded now so corrupted downloads is very likely.
2) BOINC will mark all new work with latest app version available to it.

Anything I can do to avoid download corruption other than wait for better times?

[Raistmer]

1) Probably your BOINC installation was unable to load files from berkeley's servers. Network connection overloaded now so corrupted downloads is very likely.
2) BOINC will mark all new work with latest app version available to it.

Anything I can do to avoid download corruption other than wait for better times?

Perhaps nothing It's network-related problems and if server drop connection client can't do anything with this...

[andreas]

Thanks for answering so fast.

Andreas

[Leopoldo]

Hi Andreas,

I got quite a lot compute errors withs BOINC telling me "Outpute file xy_0 for task xy is absent". On one machine I get these errors for about 1 out of 20 WU's, but the other is much worse (so I switched it to get no new work).

Output file is absent due to VLAR-killing, mod works There is no need to NoNewTasks

The 6.03 WU's from my cache were processed fine, but all new work was 6.08 WU's (prcessed fine as well).

6.03 - due to WUs already cached in project directory, BOINC marked them already with 6.03 (in client_state.xml), so BOINC will copy 6.03 app in appropriate slot directory before processing.

And all new WUs will be marked by BOINC with maximum appropriate version from app_info.xml, so 6.08

When all remaining 6.03-WUs has been processed, BOINC will delete 6.03 app from project directory.

[andreas]

Hi Andreas,

Output file is absent due to VLAR-killing, mod works There is no need to NoNewTasks

I thought I had the version without VLAR-killing, but I might have got that mixed up.

Thanks for info,
Andreas