BOINC resource allocation between 2 projects.

Message boards : Number crunching : BOINC resource allocation between 2 projects.
Message board moderation

To post messages, you must log in.

AuthorMessage
BoincSpy
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 146
Credit: 124,775,115
RAC: 353
Canada
Message 5037 - Posted: 7 Jul 2004, 15:22:59 UTC

I have my resource allocation set between 2 project:

67 % for S@Home
33 % for P@Home.

At the present time P@Home is getting about 95% of the WU's, S@home get one every 2 - three days.

I am wondering if this is an issue with S@Home limiting the number of WU's / day / user? Or could it be the resource alg is flawed.

I have a workstation that just runs S@Home and it does get WU's but only about 1 - 3 WU's at a time.

Thanks in advance.
ID: 5037 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 5047 - Posted: 7 Jul 2004, 15:36:01 UTC - in response to Message 5037.  

> I have my resource allocation set between 2 project:
>
> 67 % for S@Home
> 33 % for P@Home.
>
> At the present time P@Home is getting about 95% of the WU's, S@home get one
> every 2 - three days.
>
> I am wondering if this is an issue with S@Home limiting the number of WU's /
> day / user? Or could it be the resource alg is flawed.
>
> I have a workstation that just runs S@Home and it does get WU's but only about
> 1 - 3 WU's at a time.
>
> Thanks in advance.
>
>
There are two parts to the answer. When you first start a project, that project is behind, and you can expect to get work almost exclusively from that project until it has caught up.

Second: S@H seems to be extremely short of work at the moment (for the past couple of weeks). When a computer is refilling the queue, it doesn't find work from the project that is at the top of the list for needs work, it will move on to the next project, and it will keep moving on to the next project until it finds work or it runs out of projects. If a project does not have work, it will be contacted again at a scheduled time unless work has been before that.

ID: 5047 · Report as offensive
BoincSpy
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 146
Credit: 124,775,115
RAC: 353
Canada
Message 5051 - Posted: 7 Jul 2004, 15:46:10 UTC

Thanks John,

One question:

What determines that a project is behind? RAC / join time / WU's?

Bob

ID: 5051 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 5052 - Posted: 7 Jul 2004, 15:53:56 UTC - in response to Message 5051.  

> Thanks John,
>
> One question:
>
> What determines that a project is behind? RAC / join time / WU's?
>
> Bob
>
The resource debt which is calculated from the exp_avg_cpu (in client_state.xml) for each project, and the requested resource share. Both are calculated as a fraction of the total, an then subtracted. The largest number wins.

You can tell from the logs what the order is currently by looking to see which project gets the first connection attempt when either the low water mark is reached, or a CPU is out of work.

ID: 5052 · Report as offensive

Message boards : Number crunching : BOINC resource allocation between 2 projects.


 
©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.