Increased processing speed

Message boards : Number crunching : Increased processing speed
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile TypeOneg

Send message
Joined: 1 Aug 99
Posts: 2
Credit: 158,851
RAC: 0
United States
Message 30464 - Posted: 27 Sep 2004, 3:31:53 UTC
Last modified: 27 Sep 2004, 11:17:03 UTC

I went to this web site "http://v2.tlab404.com/" and used some of their

tweaks namely. "Increase Processing Power" currently under Top 5

Tweaks. I instituted most of the tweaks all that applied. My processing

times for Boinc-Seti has dropped signifigantly by 20 minutes on the

unit I was processing when I did the tweaks. I will give an update on

my progress in a day or two. I want to give it some time to see if it

was an anomally.
The two tweaks I did use were these.
How to Unload Cached DLL files to free memory?
How to stop "Last Access Update On NTFS?
My time on a WU has dropped to 2h 25m from 3h 5m. My times on the Seti

Classic had been in the 2h 45m range on my machine. Shuttle AN35N Ultra

Mother Board, AMD 3200+ XP, 1 Gig ram, XP Home SP2.
ID: 30464 · Report as offensive
Pascal, K G
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 2343
Credit: 150,491
RAC: 0
United States
Message 30466 - Posted: 27 Sep 2004, 4:01:08 UTC

Normal processing time can vary that much between WUWUs
My name is Pascal and this message has no meaning, but still has my approval...

It is 10 oclock, do you know what your WUWUs are doing tonight...

ID: 30466 · Report as offensive
Profile TypeOneg

Send message
Joined: 1 Aug 99
Posts: 2
Credit: 158,851
RAC: 0
United States
Message 30536 - Posted: 27 Sep 2004, 11:12:19 UTC - in response to Message 30466.  

> Normal processing time can vary that much between WUWUs
> My name is Pascal and this message has no meaning, but still has my
> approval...
>
> It is 10 oclock, do you know what your WUWUs are doing tonight...
>
>
>I updated my post with more relevant information hope that helps someone else. I know the changes have helped me. I recently started crunching Boinc-Seti and was disappointed that my times got longer by a wide margin under Boinc over Seti Classic. I don't know what might be the factors size of work units? Who knows? I am on these boards looking for help like everyone else.
ID: 30536 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 30573 - Posted: 27 Sep 2004, 15:13:26 UTC - in response to Message 30536.  

> I updated my post with more relevant information hope that helps someone
> else. I know the changes have helped me. I recently started crunching
> Boinc-Seti and was disappointed that my times got longer by a wide margin
> under Boinc over Seti Classic. I don't know what might be the factors size of
> work units? Who knows? I am on these boards looking for help like everyone
> else.

Are you saying that it takes longer to process a work unit, or that the estimated time to complete a work unit changes dramatically as the WU runs?
ID: 30573 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 30580 - Posted: 27 Sep 2004, 15:35:15 UTC

In the SETI FAQ there is a table that shows the impact of range angle on processing time.

There are potential issues with the current BOINC GUI that are not worth getting interested in for the simple reason that the cross-platform GUI will be replacing the current one anyway.

So, yes it is possible and probable that the times changed. Additional issues are things like changes in the order of processing the WU (different between Classic and BOINC powered).

It was my understanding that there were some minor changes to the processing code also.

Long term, there are efforts underway to create optimized client binaries so that we can get more effective use of the processing capabilities we have now ...
<p>
For BOINC Documentaion: Click Me!


ID: 30580 · Report as offensive

Message boards : Number crunching : Increased processing speed


 
©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.