I'm quitting Seti 'cause of increased power cost coming from "Cap and Trade"

Message boards : Number crunching : I'm quitting Seti 'cause of increased power cost coming from "Cap and Trade"
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · Next

AuthorMessage
Profile champ
Volunteer tester
Avatar

Send message
Joined: 12 Mar 03
Posts: 3642
Credit: 1,489,147
RAC: 0
Germany
Message 926359 - Posted: 15 Aug 2009, 21:34:47 UTC - in response to Message 926325.  

My wife is a graduate student working on molecular modeling for HIV. So when she isn't using her computers, I sneak into the office and run BOINC. She keeps the computers running all the time, so why not use those extra cycles? Where I am located, most of our electricity comes from nuclear, so its still relatively cheap.



But dont forget. You should run Boinc only on computers, where you got the permission to run it.


ID: 926359 · Report as offensive
Profile Byron S Goodgame
Volunteer tester
Avatar

Send message
Joined: 16 Jan 06
Posts: 1145
Credit: 3,936,993
RAC: 0
United States
Message 926428 - Posted: 16 Aug 2009, 3:54:46 UTC

Just curious for my own info since I've been thinking of limiting my own computers for costs reasons if the feature that limits cpu usage under processor usage

Use at most xxx% of cpu time

is still something that users with multiprocessor systems using 6.6.36 should not use?

This could be another limited way for me to cut back on cost if available along with

On multiprocessor system, use at most xxx% of processors

since it seems to me running at 100% cpu time also adds to the costs regardless of the percentage of the cpu's I use.
ID: 926428 · Report as offensive
Profile Gundolf Jahn

Send message
Joined: 19 Sep 00
Posts: 3184
Credit: 446,358
RAC: 0
Germany
Message 926485 - Posted: 16 Aug 2009, 8:06:22 UTC - in response to Message 926428.  

Just curious for my own info since I've been thinking of limiting my own computers for costs reasons if the feature that limits cpu usage under processor usage

Use at most xxx% of cpu time

is still something that users with multiprocessor systems using 6.6.36 should not use?

I'm not sure if anything in that direction has been done in BOINC, but the problem seems to persist, depending on the project application. Einstein@home seems to be affected mostly.

But there are other tools that achieve the task more elegantly. Look for ThreadMaster or TThrottle at the BOINC add-on software page.

Gruß,
Gundolf
Computer sind nicht alles im Leben. (Kleiner Scherz)

SETI@home classic workunits 3,758
SETI@home classic CPU time 66,520 hours
ID: 926485 · Report as offensive
Miklos M.

Send message
Joined: 5 May 99
Posts: 955
Credit: 136,115,648
RAC: 73
Hungary
Message 926533 - Posted: 16 Aug 2009, 15:51:35 UTC - in response to Message 912248.  

With my 2 P4's getting less and less credit for the wu's and the electric bill is going up it is soon not very exciting to have them running at all.
ID: 926533 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13832
Credit: 208,696,464
RAC: 304
Australia
Message 926543 - Posted: 16 Aug 2009, 17:01:23 UTC - in response to Message 926533.  

With my 2 P4's getting less and less credit for the wu's and the electric bill is going up it is soon not very exciting to have them running at all.

Don't know why they'd be gettting less & less credit unless there was something wrong with them. Since the end of June my RAC has actually increased from around 1,900 to around 2,200.
Grant
Darwin NT
ID: 926543 · Report as offensive
zpm
Volunteer tester
Avatar

Send message
Joined: 25 Apr 08
Posts: 284
Credit: 1,659,024
RAC: 0
United States
Message 926547 - Posted: 16 Aug 2009, 17:23:16 UTC - in response to Message 926543.  
Last modified: 16 Aug 2009, 17:24:37 UTC

he's means that it's taking them forever to complete a task... that another computer say a quad can knock at 2-3 times faster.. and a gpu that can knock it out 5 times faster.

I recommend Secunia PSI: http://secunia.com/vulnerability_scanning/personal/
Go Georgia Tech.
ID: 926547 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13832
Credit: 208,696,464
RAC: 304
Australia
Message 926550 - Posted: 16 Aug 2009, 17:32:44 UTC - in response to Message 926547.  

he's means that it's taking them forever to complete a task... that another computer say a quad can knock at 2-3 times faster.. and a gpu that can knock it out 5 times faster.

It's one thing for it to take longer and longer (my system is the same), it's another to be getting less & less credit.
Grant
Darwin NT
ID: 926550 · Report as offensive
Miklos M.

Send message
Joined: 5 May 99
Posts: 955
Credit: 136,115,648
RAC: 73
Hungary
Message 926609 - Posted: 16 Aug 2009, 22:27:42 UTC - in response to Message 926543.  

Recently, my RAC on one P4 dropped from 350 towards 150, judging by the credit received for the last two weeks. On my quads the credit received has gone up like yours. As far as the P4 goes it gets about 28.76 for 17,614 seconds of work.
ID: 926609 · Report as offensive
Miklos M.

Send message
Joined: 5 May 99
Posts: 955
Credit: 136,115,648
RAC: 73
Hungary
Message 926610 - Posted: 16 Aug 2009, 22:32:01 UTC - in response to Message 926547.  

Or, as it seams, an NVIDIA can complete 100 times faster.
ID: 926610 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 926648 - Posted: 16 Aug 2009, 23:47:01 UTC - in response to Message 926609.  

Recently, my RAC on one P4 dropped from 350 towards 150, judging by the credit received for the last two weeks. On my quads the credit received has gone up like yours. As far as the P4 goes it gets about 28.76 for 17,614 seconds of work.


That's the problem I have with RAC as a rating system. It doesn't reflect the amount of work a machine actually does in a given time frame unless all of the work you are doing is validation work.

This makes it rather hard to benchmark a system for seti and find the best performance per watt configuration.

I have a machine that I stopped running tasks on over a month ago, as it was just for testing, and it still gets some credit a few times a week form pending tasks.


SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 926648 · Report as offensive
Granite T. Rock

Send message
Joined: 9 Jun 99
Posts: 17
Credit: 1,248,634
RAC: 1
Canada
Message 926704 - Posted: 17 Aug 2009, 4:51:46 UTC - in response to Message 926648.  

The cost of hydro is real. I just set up my first ever dedicated machine tonight. However I did set it to stop crunching between 3 pm and 7pm which are the highest energy use times in my area. I figure there's no point adding to the strain of the grid. However... I do have the windows update run during that time. Too bad I can't program my PC to auto start-up at certain times.
ID: 926704 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13832
Credit: 208,696,464
RAC: 304
Australia
Message 926721 - Posted: 17 Aug 2009, 7:03:38 UTC - in response to Message 926609.  

Recently, my RAC on one P4 dropped from 350 towards 150, judging by the credit received for the last two weeks.

What about pending credit?
With longer processing times, on slower machine RAC will drop in the short term, but continue to rise in the longer term (till it gets to what's the general "average"). Pending credit will climb sharply.
If after a month or so the RAC hasn't come back up to roughly where it was- there is a problem with your machine no longer processing as much work as it used to.
Grant
Darwin NT
ID: 926721 · Report as offensive
Rasputin
Volunteer tester

Send message
Joined: 13 Jun 02
Posts: 1764
Credit: 6,132,221
RAC: 0
Russia
Message 926733 - Posted: 17 Aug 2009, 9:57:10 UTC - in response to Message 926704.  

Too bad I can't program my PC to auto start-up at certain times.


You can if your bios supports it. Check your bios for a "wake on alarm" or "power on by alarm" setting.
ID: 926733 · Report as offensive
Miklos M.

Send message
Joined: 5 May 99
Posts: 955
Credit: 136,115,648
RAC: 73
Hungary
Message 926743 - Posted: 17 Aug 2009, 13:01:16 UTC - in response to Message 926721.  

Thank you Grant, I will keep an eye on the RAC before i retire it.
ID: 926743 · Report as offensive
Profile ccappel
Avatar

Send message
Joined: 27 Jan 00
Posts: 362
Credit: 1,516,412
RAC: 0
United States
Message 926744 - Posted: 17 Aug 2009, 13:33:57 UTC - in response to Message 926609.  
Last modified: 17 Aug 2009, 14:29:26 UTC

Recently, my RAC on one P4 dropped from 350 towards 150, judging by the credit received for the last two weeks. On my quads the credit received has gone up like yours. As far as the P4 goes it gets about 28.76 for 17,614 seconds of work.

I think I see where he is coming from. The recent rash of shorties has lowered his CPU seconds per credit on his P4. A 92 credit WU was taking on average 43,000 CPU seconds, which comes out to under 500 CPU seconds per credit, while the shorties are getting 28.75 credit in roughly 17,500 CPU secs...an average of over 600 CPU seconds per credit.

EDIT: That doesn't account for the 50% drop in RAC for that box, but still...

I just checked my Core2Duo and my ratios are 97 and 83, respectively, so the shorties have actually improved my CPUsec/credit ratio a bit.
"Life is a tragedy for those who feel, and a comedy for those who think."

"I never get into an argument that I cannot win."
ID: 926744 · Report as offensive
Miklos M.

Send message
Joined: 5 May 99
Posts: 955
Credit: 136,115,648
RAC: 73
Hungary
Message 926791 - Posted: 17 Aug 2009, 17:24:26 UTC - in response to Message 926744.  

I wonder if my P4 is running hotter than before and thus less efficient?
ID: 926791 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 926798 - Posted: 17 Aug 2009, 17:55:43 UTC - in response to Message 926791.  

It's entirely possible. Dust buildup around the fans can prevent them from moving the heat fast enough, thus causing your CPUs to run hotter and less efficient.

Summertime heat can also have an effect if the ambient room temperature is rather warm.
ID: 926798 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 926806 - Posted: 17 Aug 2009, 18:20:48 UTC - in response to Message 926798.  

It's entirely possible. Dust buildup around the fans can prevent them from moving the heat fast enough, thus causing your CPUs to run hotter and less efficient.

Summertime heat can also have an effect if the ambient room temperature is rather warm.


I changed out a fan on an older system last night as it had started to make a lot of noise. After pulling the fan off I then pulled the dust off of the heat sink in a mat. It was both funny and disturbing. I have to remember to clean out my older machines as often as I do my main one.

As I figured I'd clean other systems while I was at it. My MODT server, which I pretty much haven't opened up since I built it, actually had cobwebs inside of the case.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 926806 · Report as offensive
zpm
Volunteer tester
Avatar

Send message
Joined: 25 Apr 08
Posts: 284
Credit: 1,659,024
RAC: 0
United States
Message 926828 - Posted: 17 Aug 2009, 20:26:39 UTC - in response to Message 926806.  

It's entirely possible. Dust buildup around the fans can prevent them from moving the heat fast enough, thus causing your CPUs to run hotter and less efficient.

Summertime heat can also have an effect if the ambient room temperature is rather warm.


I changed out a fan on an older system last night as it had started to make a lot of noise. After pulling the fan off I then pulled the dust off of the heat sink in a mat. It was both funny and disturbing. I have to remember to clean out my older machines as often as I do my main one.

As I figured I'd clean other systems while I was at it. My MODT server, which I pretty much haven't opened up since I built it, actually had cobwebs inside of the case.



YUK!! you mean a spider has been living in your server... MY GOD

I recommend Secunia PSI: http://secunia.com/vulnerability_scanning/personal/
Go Georgia Tech.
ID: 926828 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 926843 - Posted: 17 Aug 2009, 21:19:00 UTC - in response to Message 926704.  

Too bad I can't program my PC to auto start-up at certain times.

Why not?

Many modern BIOSes have settings to power on at a specific time, and most operating systems have scheduler or "chron" daemons that allow programs or events to be run by schedule.
ID: 926843 · Report as offensive
Previous · 1 · 2 · 3 · 4 · Next

Message boards : Number crunching : I'm quitting Seti 'cause of increased power cost coming from "Cap and Trade"


 
©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.