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

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 7
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 7191 times and has 6 replies Next Thread
gasteed
Cruncher
Joined: Mar 31, 2009
Post Count: 2
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Mapping Cancer causes basic machine lockup on multi-processor system

Hi,

I've been a WCG since 3/09 and have run it on a variety of machines. My current machine is a 6 core amd 1075t. My issue is that when the mapping cancer runs it basically takes over my HDD virtual memory and my pc becomes a brick. No other wcg project has caused this issue. Running win7 home premium 64 bit. Question is, I'd like to run the mapping cancer (dear friend surviving Leukemia) but I still need to use my machine. Is there a way to have it run only one instance as to the 5 (which is when my pc became a brick) instances? It's a six core and I let it run 100% cpu usage 24/7 with no issues. But if the mapping kicks in with more than one running at a time my pc becomes useless, lol.

TIA Gordon...
[Nov 22, 2013 8:49:16 PM]   Link   Report threatening or abusive post: please login first  Go to top 
sgoll
Advanced Cruncher
Joined: Oct 24, 2006
Post Count: 87
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Mapping Cancer causes basic machine lockup on multi-processor system

In the boinc manager window you can go to:
-> Tools -> Computing preferences -> processor usage -> On multiprocessor systems ...
There you can define how many cpu cores you will spent for boinc. For 3 of 6 cores you should select 50%. You can play around with this values to find a comfortable setting.
Stephan
----------------------------------------

[Nov 22, 2013 10:52:24 PM]   Link   Report threatening or abusive post: please login first  Go to top 
OldChap
Veteran Cruncher
UK
Joined: Jun 5, 2009
Post Count: 978
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Mapping Cancer causes basic machine lockup on multi-processor system

Gordon,

How much memory do you have in that machine?

If you go to tools>Boinc preferences>Disk and memory usage: what percentage of memory is showing? What percentage of the swap file too?
----------------------------------------

[Nov 22, 2013 11:16:01 PM]   Link   Report threatening or abusive post: please login first  Go to top 
ccandido
Senior Cruncher
Joined: Jun 22, 2011
Post Count: 182
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Mapping Cancer causes basic machine lockup on multi-processor system

If you write an app_config.xml and put it in the boincdata folder of the wcg project you can run one wu of the Mapping Cancer project and as many wu of any other project.

for instance, i have this to limit to one the number of lattice wu

<app_config>
<app>
<name>garli</name>
<max_concurrent>1</max_concurrent>
</app>
</app_config>

what you need to do is to replace the name of the aplication which you can find in your client_state.xml.

When you open that file (with note pad) be extremely careful not to change anything.
----------------------------------------


[Nov 22, 2013 11:38:24 PM]   Link   Report threatening or abusive post: please login first  Go to top 
gasteed
Cruncher
Joined: Mar 31, 2009
Post Count: 2
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Mapping Cancer causes basic machine lockup on multi-processor system

Thanks ccandido, That's exactly what I was looking for as I wanted to continue to allow all 6 processors to crunch wu, just not multiple mappings. I will give this a try, and give status if I can make it work....
[Nov 23, 2013 12:39:30 PM]   Link   Report threatening or abusive post: please login first  Go to top 
OldChap
Veteran Cruncher
UK
Joined: Jun 5, 2009
Post Count: 978
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Mapping Cancer causes basic machine lockup on multi-processor system

I can't help feeling something else needs adjusting, I run a rig with half a GB ram per core, less what OS uses, with no issues on MCM1
----------------------------------------

[Nov 23, 2013 12:48:19 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: Mapping Cancer causes basic machine lockup on multi-processor system

Once one gets past the point of what the OS uses, 0.5GB per MCM would easily handle 6 [probably 2GB for 6 cores could even do things, since the big search models are more of a rarity] My quad, MCM only, lists 7.26 versions since launch in range of peak RAM of 10 to 64MB, and VM from 15 to 103MB. Even if all 4 peaked simultaneous with these figures, it's possible to run 4 in 1GB allocated RAM.
[Nov 23, 2013 1:10:22 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread