Folding@Home

RTmarc

There has been some intra-office discussion over the past few days regarding the possibility of including the Folding@Home application in our new image. We have begun the process of creating new images (ideally) every quarter to push out to our workstations. Essentially we would have all of the workstations on the new images part of a single team contributing to the cause. I was curious if anyone else was doing this or have done this in the past.

For those of you who have never heard of of Folding@Home, think distributed computing similar to SETI@Home but for a good cause. Folding is targeted towards protein folding and manipulation.


Oh, and here is a link to their website: http://folding.stanford.edu/


EDIT: This image is typically only deployed as needed (system failure or new machine purchase) so the application would not be loaded on archaic machines. Nearly every machine we have ordered within the past year has been 2.5+GHz Dual-Core Intel with 2GB of RAM. There are a few others that are slightly more robust than this (including mine ) that are already running the app.

astorrs

Make sure you have buy in (in writing) from the head of IT.

RTmarc

astorrs wrote:

Make sure you have buy in (in writing) from the head of IT.

That would be me.

astorrs

Then get the okay from your boss.

eMeS

RTmarc wrote:

There has been some intra-office discussion over the past few days regarding the possibility of including the Folding@Home application in our new image. We have begun the process of creating new images (ideally) every quarter to push out to our workstations. Essentially we would have all of the workstations on the new images part of a single team contributing to the cause. I was curious if anyone else was doing this or have done this in the past.

EDIT: This image is typically only deployed as needed (system failure or new machine purchase) so the application would not be loaded on archaic machines. Nearly every machine we have ordered within the past year has been 2.5+GHz Dual-Core Intel with 2GB of RAM. There are a few others that are slightly more robust than this (including mine ) that are already running the app.

This is just my opinion...so please take it as nothing more than that. There are arguments on all sides of this issue, and it could be debated ad infinitum.

First, I'm assuming that you work for a for-profit business? If that is the case, doesn't it make more sense for workstations to go into hibernate mode to conserve energy when not in use? Or, have you thought about claiming some kind of tax deduction for the energy and processing power you'll be donating to FAH (basic research)? The cost of running FAH could be significant depending on the size of your environment.

For example, if I were to leave my PS3 on all the time running FAH, the cost was between $200 and $250 per year. Not much relative to other things, but I choose to direct my charitable efforts differently. (I have a device called a "Kill-A-Watt" that lets me measure how much energy a device uses over time).

Second, I'm of the opinion that a production release of any software or o/s should only contain exactly what is needed to conduct the business at hand. My opinion is based on the fact that the simplest solution with the fewest complications is best. Any unnecessary things added to a package, release, or image are the potentially the cause of future problems.

Not that FAH will necessarily cause problems, but why even allow the potential for problems to occur by installing an application into an image that isn't needed to conduct business? I wouldn't want to be the one place that finds a potential single interaction between FAH and some random software update that shuts down the business....

RTmarc wrote:

For those of you who have never heard of of Folding@Home, think distributed computing similar to SETI@Home but for a good cause. Folding is targeted towards protein folding and manipulation.

This implies that SETI@Home is not a good cause. The fact is that both SETI@Home and FAH are both conducting basic research. Good or bad is irrelevant as the intent is to expand knowledge. It is impossible to say at this point whether any of it is "for a good cause" or if any of this research will result in anything that ever becomes applied.

Academically interesting? Yes. However, IMO if any of this were likely to produce a worthwhile result then we would see some kind of commercial activity driving this (There are some notable exceptions here), rather than these programs depending on grants and donations.

IMO, it will be a while before private equity drives something like this again (on a large scale), given what happened to Celera with the human genome project, when Clinton declared in 2000 that the decoded genome could not be patented. Biotech in general lost about $50 billion in market cap in two days when this happened. Oddly enough, this occurred right inline with the beginning of the dot com bust.

MS

RTmarc

We are a non-profit organization. Also, machines are left on at night for patching, virus/vulnerability scans, software package deployments, etc so we wouldn't really be losing anything there. We do have a policy that monitors, printers, and any non-essential peripherals are turned off at night to conserve power.

As far as environment goes we're in the neighborhood of 125-140 workstations and 35 servers.

Thanks for the input!

eMeS

RTmarc wrote:

We are a non-profit organization. Also, machines are left on at night for patching, virus/vulnerability scans, software package deployments, etc so we wouldn't really be losing anything there. We do have a policy that monitors, printers, and any non-essential peripherals are turned off at night to conserve power.

As far as environment goes we're in the neighborhood of 125-140 workstations and 35 servers.

Thanks for the input!

Ah, I assumed incorrectly!

I could see FAH being more of a fit in a non-profit environment, especially one that focuses on one specific malady or disease group. Why not roll the dice for basic research that might one day help understand something that is the focus of the organization, especially since you've indicated there is no additional cost incurred.....

It sounds like this is in your budget, and you're the decision maker, so you have all that you need. My only concern would be (as mentioned before) the risk of having software installed that is not critical to the organization's business potentially affecting software that is critical to the organization's business.

I know of many organizations that check workstations when the user signs on for unauthorized software. At one specific company I know that if something is found, a fresh image is automatically deployed on the offending workstation! This is how they address the risk that someone is installing something that could impact their business.

MS