Results 1 to 4 of 4

Thread: Distributed Computing Wattage Monitoring

  1. #1
    Senior Member
    Join Date
    Dec 2001
    Posts
    884

    Distributed Computing Wattage Monitoring

    Heya folks.

    I and a few others have just completed a project where we tried to figure out the power usage of distributed.net's dnetc application when running on your computer, so that people could figure out how much it "costs" to run it (besides the obvious wear on your computer's processor).

    The distributed.net application, in case you don't know, is a client that essentially crunches numbers on your computer in small packets, and sends these back to a server that analyzes the outcome. This is done on a mass scale, hence the word "distributed." This is sort of similar to other projects such as SETI@Home and United Devices Cancer Research (grid.org). The goal of the project(s) can be read more about on distributed.net's website.

    So, it is now posted over at ReadyResponse for any of you to read if you like. I'd have to say it's a good read (I wrote it, how could it not be? :-p), and would appreciate any and all feedback from you all.

    Link to the tutorial: http://www.readyresponse.org/index.p...pt=6&Itemid=52

    Small excerpt:
    This article will serve as an informational reference for those of you who are interested in finding out exactly how much power usage you’re using by running distributed.net’s client on your computer. We chose the distributed.net RC5 clients because it would require 100% CPU usage. Most other distributed projects require the same workload and, as such, were not tested.

    ...

    There were various different computer processors that were used in this analysis. There was a PowerMac G5 dual-processor 2 ghz, a P3 800 mhz, and a P4 1.8 ghz. Not surprisingly, each of these setups gave different power usage patterns, which you can read about in more detail in the following paragraphs.
    Hope you all enjoy.

  2. #2
    Senior Member
    Join Date
    Oct 2001
    Posts
    114
    Now this is something that's very interesting.

    Although you did not test AMD's but I will expect them to be equivalent to P4's. I am glad to know that I am using only 40-51w/hr extra by running dnet. Gosh I used to be scared.

    This turns out to what..by the rate of 12cents/KWhr (after conversion to USD), and average usage of 50w/hr...

    12 cents for 20 hours..
    14.4 cents for a day.
    4.32 dollars a month..

    wow suddenly thats too much
    Better Laugh At Your Own Problems..
    Coz...The World Laughs At Them

  3. #3
    Senior Member
    Join Date
    Dec 2001
    Posts
    884
    I'm glad ya like it. Keep in mind, if you or anyone else is interested in letting us use statistics that they gather using their own boxes, we're definitely open to incorporating the new numbers into a follow-up paper. Thanks.

  4. #4
    Senior Member RoadClosed's Avatar
    Join Date
    Jun 2003
    Posts
    3,834
    Nice Article. Basically its like leaving the front and back porch light on. Assuming each bulb is 150 watts. Here is an interesting correlation. At idle the PC uses the same energy as say, leaving the living room light on. Most of us would think, what a waste. Then add in the other computers, stereo with it's lights on, VCR, TV in standbye, DVD etc and it adds up to quite a bit. Those things aren't "off" most of the time.
    West of House
    You are standing in an open field west of a white house, with a boarded front door.
    There is a small mailbox here.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •