Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 198
Posts: 198   Pages: 20   [ Previous Page | 1 2 3 4 5 6 7 8 9 10 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 30623 times and has 197 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Hello,

My config:
- windows 7 64 bits
- 2 gtx 580
- i7 980x

I want to work more of one task per GPU at same time.

I try to put "app_info.xml" fin in the worldcommunitygrid folder.
----------
<app_info>
<app>
<name>hcc1</name>
<user_friendly_name>Help Conquer Cancer</user_friendly_name>
</app>
<file_info>
<name>wcg_hcc1_img_7.05_windows_intelx86__nvidia_hcc1</name>
<executable/>
</file_info>
<file_info>
<name>hcckernel.cl.7.05</name>
<executable/>
</file_info>
<app_version>
<app_name>hcc1</app_name>
<version_num>705</version_num>
<platform>windows_intelx86</platform>
<plan_class>nvidia_hcc1</plan_class>
<avg_ncpus>1.0</avg_ncpus>
<max_ncpus>1.0</max_ncpus>
<coproc>
<type>NVIDIA</type>
<count>.50</count>
</coproc>
<file_ref>
<file_name>wcg_hcc1_img_7.05_windows_intelx86__nvidia_hcc1</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>
----------

Now i receive no task and i have errors messages:
----------
12.11.2012 19:45:27 World Community Grid Requesting new tasks for CPU
12.11.2012 19:45:30 World Community Grid Scheduler request completed: got 1 new tasks
12.11.2012 19:45:30 World Community Grid [error] App version returned from anonymous platform project; ignoring
12.11.2012 19:45:30 World Community Grid [error] No application found for task: windows_intelx86 705 ; discarding
12.11.2012 19:46:45 World Community Grid Sending scheduler request: To fetch work.
----------

Can you help me please ?

Sorry for my English it's not my first language.
[Nov 12, 2012 6:50:33 PM]   Link   Report threatening or abusive post: please login first  Go to top 
rodgers82
Cruncher
Joined: Jul 13, 2009
Post Count: 7
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Problem solved!! Thank you very much guys.

Detaching and reattaching the project was the solution.

I could not detach because I was using BAM!

now it's downloading task.
[Nov 13, 2012 1:45:30 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Platoon
Advanced Cruncher
Russia
Joined: Jun 28, 2006
Post Count: 62
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

I am looking for advice how to optimize GPU/CPU crunching, my config is:
- 980X with HT - 12 threads
- ATI 7950 at 10 threads => 10 GPU tasks + 2 HCC CPU task
- Win7-64
- boinc 7.05

I am thinking about 12 threads on GPU at cost of 2 CPU tasks, but not sure if there will be no bottleneck.
Don't want to do experiments - any advise how to improve system output?
Thank's!
----------------------------------------
" forever forge ahead and keep the dream in sight!"

----------------------------------------
[Edit 1 times, last edit by Platoon at Nov 13, 2012 7:37:32 AM]
[Nov 13, 2012 7:36:27 AM]   Link   Report threatening or abusive post: please login first  Go to top 
ThreadRipper
Veteran Cruncher
Sweden
Joined: Apr 26, 2007
Post Count: 1320
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

The best solution would be a WCG website update where you could choose how many simultaneous HCC GPU WUs you would like to run at the same time - defaulting to 1 of course. That way, "power users" could change it while everyone else could run with no problems...much the way CEP2 is handled today (even though it does not specify parallel tasks, just how many are downloaded in the same WU cache). But almost.
----------------------------------------

Join The International Team: https://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=CK9RP1BKX1

AMD TR2990WX @ PBO, 64GB Quad 3200MHz 14-17-17-17-1T, RX6900XT @ Stock
AMD 3800X @ PBO
AMD 2700X @ 4GHz
[Nov 13, 2012 7:49:26 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Very not likely, simple because the client / server structure to do so does not exist, and imagine a user having painstakingly created an app_info file, makes a change in the web profile, and poof, gone. you'd be needing similar to the global_prefs + global_prefs_override, 2 files, app_info.xml and app_info_override.xml. Speak to the developers ;>)
[Nov 13, 2012 12:13:50 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Ingleside
Veteran Cruncher
Norway
Joined: Nov 19, 2005
Post Count: 974
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Very not likely, simple because the client / server structure to do so does not exist

Einstein@home already has the option for users to specify #task per GPU, without using app_info.xml at all...

So, if WCG wants to do something similar, it's just to talk to Einstein@home.
----------------------------------------


"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."
[Nov 13, 2012 6:25:56 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Frankly, what's the gain? You get an "In progress" filter, but at application level. Still does not tell the GPU that they can process X concurrent, or does it?
[Nov 13, 2012 6:33:58 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Ingleside
Veteran Cruncher
Norway
Joined: Nov 19, 2005
Post Count: 974
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Frankly, what's the gain? You get an "In progress" filter, but at application level. Still does not tell the GPU that they can process X concurrent, or does it?

Of course it's to crunch X concurrently on a GPU.
----------------------------------------


"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."
[Nov 13, 2012 6:58:16 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Not sure if this is the one, but don't think there's need to talk to Einstein on this for WCG:
GPU utilization factor of BRP apps: 1
DANGEROUS! Only touch this if you are absolutely sure of what you are doing!
Wrong setting might even damage your computer! Use solely on your own risk!
Min: 0.0 / Max: 1.0 / Default: 1.0

Somehow think it's most improbable to get a WCG device profile slot.
[Nov 13, 2012 7:17:17 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: GPU Optimisations

Because Uplinger already stated that WCG's official stand is that there is no WCG official support for app_info file, users who are contemplating to use the app_info (mainly for customized GPU use in crunching WCG WUs) -- are on their own -- sad as that may be.

To those crunchers who want to venture off WCG-official-support and do app_info anyways, here's an informative PDF link* about the inner-workings of app_info for use here at WCG.

Notes:
*extracted from OldChap [Nov 12, 2012 11:05:37 PM] post
;
[Nov 13, 2012 10:22:44 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 198   Pages: 20   [ Previous Page | 1 2 3 4 5 6 7 8 9 10 | Next Page ]
[ Jump to Last Post ]
Post new Thread