Seti Preeminence in Productivity?

Message boards : Number crunching : Seti Preeminence in Productivity?
Message board moderation

To post messages, you must log in.

AuthorMessage
PhonAcq

Send message
Joined: 14 Apr 01
Posts: 1656
Credit: 30,634,091
RAC: 293
United States
Message 1915475 - Posted: 27 Jan 2018, 19:59:24 UTC

I notice that we/SETI@Home has dropped to fifth place in productivity (BoincStats). I recall that once we were probably #1 or there about. Is it worth asking how this project can be made more productive and regain its preeminence?
ID: 1915475 · Report as offensive
Profile Wiggo "Socialist"
Avatar

Send message
Joined: 24 Jan 00
Posts: 14932
Credit: 193,628,352
RAC: 54,207
Australia
Message 1915477 - Posted: 27 Jan 2018, 20:11:50 UTC
Last modified: 27 Jan 2018, 20:12:05 UTC

CreditNew (or CreditScrew) is to blame for that and you can read more about that here.

Cheers.
ID: 1915477 · Report as offensive
PhonAcq

Send message
Joined: 14 Apr 01
Posts: 1656
Credit: 30,634,091
RAC: 293
United States
Message 1915484 - Posted: 27 Jan 2018, 20:41:19 UTC - in response to Message 1915477.  

Is this item, creditscrew(?), applied uniformly across BOINC projects? Is there a short synopsis of why it was implemented? Is there chaos at the helm of this ship?
ID: 1915484 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 5567
Credit: 386,073,999
RAC: 1,034,004
United States
Message 1915488 - Posted: 27 Jan 2018, 20:45:22 UTC - in response to Message 1915484.  
Last modified: 27 Jan 2018, 20:47:17 UTC

It is if they use the stock BOINC server software and the CreditNew algorithm for assigning task credit. Any project can either adopt the stock credit mechanism or abandon it and apply their own credit valuations. Many projects do that and assign much greater credit for equivalent flops of processing.

You can read the CreditNew Wiki here
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1915488 · Report as offensive
PhonAcq

Send message
Joined: 14 Apr 01
Posts: 1656
Credit: 30,634,091
RAC: 293
United States
Message 1915490 - Posted: 27 Jan 2018, 20:50:52 UTC - in response to Message 1915488.  

I thought this "standardization of credit between projects" problem was solved years ago. Perhaps I don't remember correctly. I've been doing this project for I guess 17 years (yikes!) so I don't remember everything, I guess. But, without standardization, what good is BoincStats or other tools to compare projects or historical performance within a project? This doesn't sound very scientific to me!!
ID: 1915490 · Report as offensive
Profile Wiggo "Socialist"
Avatar

Send message
Joined: 24 Jan 00
Posts: 14932
Credit: 193,628,352
RAC: 54,207
Australia
Message 1915491 - Posted: 27 Jan 2018, 20:53:27 UTC - in response to Message 1915484.  

Is this item, creditscrew(?), applied uniformly across BOINC projects? Is there a short synopsis of why it was implemented? Is there chaos at the helm of this ship?

No it's not as many projects at the time of introduction saw the flaw in it and those projects out running us now did not pick the flawed concept up.

Cheers.
ID: 1915491 · Report as offensive
PhonAcq

Send message
Joined: 14 Apr 01
Posts: 1656
Credit: 30,634,091
RAC: 293
United States
Message 1915492 - Posted: 27 Jan 2018, 20:53:36 UTC - in response to Message 1915488.  

Man, Keith! You have a lot of compute power. Impressive.
ID: 1915492 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 5567
Credit: 386,073,999
RAC: 1,034,004
United States
Message 1915496 - Posted: 27 Jan 2018, 21:23:04 UTC - in response to Message 1915492.  

I don't have the hundreds of hosts farms like some in the Top 20, but my four PC hosts do alright. Biggest improvement was adopting the Linux special app.
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1915496 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 5567
Credit: 386,073,999
RAC: 1,034,004
United States
Message 1915497 - Posted: 27 Jan 2018, 21:27:03 UTC - in response to Message 1915490.  

I thought this "standardization of credit between projects" problem was solved years ago. Perhaps I don't remember correctly. I've been doing this project for I guess 17 years (yikes!) so I don't remember everything, I guess. But, without standardization, what good is BoincStats or other tools to compare projects or historical performance within a project? This doesn't sound very scientific to me!!

Well, there really is no standardization. But I still use BoincTasks for quick overviews of my daily host production. You can see the inequalities in credit valuations among projects here
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1915497 · Report as offensive
PhonAcq

Send message
Joined: 14 Apr 01
Posts: 1656
Credit: 30,634,091
RAC: 293
United States
Message 1915680 - Posted: 28 Jan 2018, 23:20:28 UTC

This situation is kind of disillusioning. I've joined a prime number search project to co-compute with SETI for awhile as a result.
ID: 1915680 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 5567
Credit: 386,073,999
RAC: 1,034,004
United States
Message 1915688 - Posted: 29 Jan 2018, 0:19:29 UTC - in response to Message 1915680.  

This situation is kind of disillusioning. I've joined a prime number search project to co-compute with SETI for awhile as a result.

And that is going to happen a lot more with the growing disillusionment with the project's credit system. We are going to lose ever more hosts that move to other projects that award credit sanely.

I know that I have joined other projects and have reduced my resource share of Seti to move share to other projects. Just to try and keep my RAC steady or not drop as precipitously.
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1915688 · Report as offensive

Message boards : Number crunching : Seti Preeminence in Productivity?


 
©2018 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.