Ever heard of Distributed Computing? Well, I had looked into it a while ago but I just wasn’t sure if I wanted to do it. Recently I thought about how much downtime my computer has (when I’m sleeping, at school, at work, out climbing, out camping, etc.) and it turns out it’s a lot – most of the time in fact. I still don’t feel like shutting it off ever, so it ends up not doing anything while it’s on except wait for me to return. I finally decided get involved with distributed computing and it’s actually pretty cool. I feel like I’m doing something good.
My computer is part of Stanford’s Folding at Home project which simulates the folding of proteins, RNA and nanoscale synthetic polymers in an attempt to discover more about Alzheimer’s Disease, Huntington’s Disease, Cancer, Osteogenesis Imperfecta, Parkinson’s Disease and the Ribosome. A worthy cause indeed!
If you get the graphical client for Windows, this is what it will look like when it’s running:
You can, of course, turn this off and just let the client run in peace. You can change a ton of settings and run the client as you like it. You can, for example, set it to run all the time at the lowest priority – that way if any other program needs more cpu time, FAH automatically backs down. You can set how much processor you want to allow FAH to use at most, I keep mine usually at about 90 percent. You can also set FAH to run only when it’s in screensaver mode. This means you set the screensaver to turn on after x amount of time and the FAH client does not use any of your resources unless the computer has been idle for that time.
When you open up the client you see the protein or fold currently being worked on. You can set it to display as the ball and stick model or the space-fill model. The information on the left shows your your name, your team number (446 is the google team), the number of WU (work units) you’ve finished, protein that’s being worked on, how many frames have being simulated out of the total and average time it takes to process one frame. Below that it shows when the next frame will be done and below that it shows how long it will take, approximately, to finish the entire work unit. This will change depending on how much free processor time the FAH client can get.
If you’re interested in distributed computing, visit http://folding.stanford.edu/ for more information.
