My computer is my space heater sometimes, and currently my room is being heated by Project 14328 simulating the main COVID-19 protease and receptor binding domains.
Join me on team 243401.
13 Likes
On my 8 chips and gpus I’ve done 84 Work Units so far. Feel free to join team 243401 !
5 Likes
Currently our team stands at 950,000 points. Can you help us break one million? foldingathome.org Team 243401
4 Likes
Joined the team pixl.us in Folding@Home
Not sure if I’m contributing to the same research project as the one set up by the team, though…
1 Like
Oh neat! I tried to setup a desktop to join but there’s a firewall appliance in the way, and getting a darn SOCKS + HTTP proxy setup to work with F@H has been a real PITA so far. So I had to stop working on it for a bit.
I may come back and look at it again in further detail soon and see if we can get it working. Sorry!
1 Like
Awesome~ In that drop down you should be able to select COVID-19 (this is a new version of the client, previously you would just select “Any”)
Currently you are simulating one of the polymerase proteins of the COVID-19 causing SARS-CoV-2 coronavirus! This is a potential target for a drug which would prevent this coronavirus from reading it’s own data.
Projects 16425
Cause: covid-19
This project simulates the NSP7 co-factor in the monomeric state from SARS-CoV-2! To replicate itself, the virus uses “polymerase” proteins that transcribe the viral genome after it infects a host cell (aka our cells). The known polymerase of the viruses utilizes 3 proteins, NSP7 and NSP8, and NSP12. This project simulates NSP7 in isolation. Our hope is to identify a potentially druggable site in this polymerase for drug dessign efforts. This project is related to p16424, p16431, and p16432.
1 Like
Thanks to new team members
@gadolf @heckflosse @gaaned92 Supertobi and an Anonymous donor…
We have broken a million points!
1,025,537
Current stats: Folding@home | Statistics
Energy usage note: Each of my three desktop systems are working pretty much continuously. The draw is probably around 100W each. According to the national average (tiered usage), it is costing me $50 per month to run the protein calculations. Not that we have that kind of money laying around, but there is an altruism at play. It is the personal sacrifice I am willing and demanding of myself to make because I am a believer in computer simulations, having designed/built/run compute clusters for electro-magnetics problems. In my heart, and in the heart of most folding@home donors, we know that scientists need to have the power to quickly determine if a question is right or wrong. In my past work, it was even at times to prove an engineer wrong, requiring the exactitude of software that costs annually as much as the United States national median housing cost. The folding@home network is now more powerful than the top 500 supercomputers combined. Ten years ago, most engineers would look at you sideways if you told them you used simulators to solve problems.
2 Likes
I’ve been thinking about joining several times. However, I could not make up my mind because the software (neither client nor server side) is not open source. To what I could figure out, only the GUI front-ends are… Please, let me know if I’m wrong here.
This poses some trust issues (having a software using all my PC resources, without the chance to know what it’s doing, namely); not saying that I don’t trust this project in particular, but it’s a principle.
Is there a real valid reason for that?
Isn’t it a commercial/copyright protection reason, mainly (a part from the trust one)? (and in that case, it could be like taking advantage of free contribution).
I know the matter is sensitive and it’s hard to argue against a good cause… but still… I’d like to solve this personal conundrum, if anybody more familiar with the matter could shine some light.
Thanks!
EDIT:
PS: main sources of my info about their approach to OS/non-OS:
and the F@H GitHub: Folding@home · GitHub
1 Like
That’s certainly amazing.
On the power consumption side, though, I’d like to add a slight negative remark (not that I don’t appreciate the effort, of course, but there’s no light without shadow… what matters is the balance).
Fold@Home uses systems in a highly non-optimized way, energy-wise (think about the energy spent in communication alone, across all the net) to perform what it’s doing, so in comparison to specifically designed calculus centers, the environmental impact (per bit of result, let’s say) it’s much higher.
The CO2-equivalent production of the Fold@Home net is much higher, per result, compared to optimized computer farms (at least, until energy is largely produced by fossil fuel, not only at your home, where you may be able to control the energy source, but on all the network that carries your produced data).
This being said, it’s not on me, of course, to say if this is worth or not the environmental impact, or if we’d be much better- or worse-off without it.
I think it’s important to always see all the aspects, though. Sorry if it seems I rain on your parade, it’s not at all my intent.
I believe it is for security as people in the past have tried to cheat on the points system. Cheating with the client software ruins the scientific integrity of the study. Network communications occur for a few seconds per work unit, which take a couple of hours to a couple of days, and you don’t need to run thousands of GPU cores to send a work unit back to the university. The real energy hogs, gromacs and openmm, are developed open-source and FAH has a special license grant to benefit from that in their closed-source, for scientific integrity. They are the molecular simulators. To me that seems reasonable, because it is for accuracy’s sake. There is a viewer which animates the results of your calculations during each run. (Before any results come out of your simulation it displays a demonstration molecule). Scientific information about this coronavirus is in short supply, and we want to avoid a situation where we just “try” stuff for months and months with no guarantee of hitting the target. If and when a vaccine is developed, it’s going to go through a fair amount of controlled testing which will take a long time, and sadly people will expire in the duration. The results of F@H intend to accelerate this, hence the source-code exemption grant. Lives are at stake, be it Cancer, Alzheimers, or Covid-19, in significantly critical numbers.
There are better reasons besides the open/closed source issue for which you would not run F@H. For example, you could melt down your computer/laptop fairly quickly if the thermal handling is sub-optimal. And if you are trying to keep a roof over your head using precious computing resources, you certainly wouldn’t want that happening.
Thanks for taking the time and addressing my concerns.
I’m still not convinced but I understood your explanation. However, when it comes to scientific integrity, you can still enforce that the computing is performed by a valid instance of a client without the need to close the source.
For the point system, I honestly don’t know. Is there a reward mechanism? In that I could see how any system involving trade mechanisms is more easily (maybe?) managed through closed software.
Anyhow, I’m not saying this is not a great tool and it’s not helping. At all. I’m definitely skeptical by nature, I guess… But I don’t want to convince anybody to stay away from it. But I felt this is the place to share my concerns.
As for the energy discussion, I don’t want to convince anybody. And it’s also not the point, I guess. I’m pretty sure of what I know and understand of the subject, but it’s also a matter quite delicate and risky to discuss. It is a trade-off that should not be ignored, though.
I’ll think about it a bit more… 
Thank you for the patience in listening to me and answering. I truly appreciate.
1 Like
Well, you have to remember the first wide-scale deployment of F@H occurred on the 2007 PlayStation3 network. The reward of the point system is just the points, as if you were going for the high score. If you played a PS3 multi-player network game like Warhawk, you remember cheating was rampant, and for the sole purpose of attaining the ultimate hack, or some high rank.
The PS3 offered something more than a plain-ole’ CPU, and F@H took advantage of that, which led to development of the GPU calculator. Alzheimers research entered a new phase of understanding thanks to the PS3, which was retired by F@H in 2012.
You are right to be skeptical, that is a very scientific attitude to have.
3 Likes
Today I’m switching the hardware support, from a 24x7 windows desktop to a 24x7 ubuntu server. I installed Cacti to keep monitoring its health.
While checking team stats, I was curious to go to the #1 team’ web page and I got confused: it seems that anyone who joins that team is eligible to receive a type of bitcoin (curecoin) in return, based on some kind of complex algo. So people are giving their computer power in exchange for … money?? (In the end, I think I still don’t understand what’s this bitcoin fuss about )
I have no clue. I’m in it purely for the vaccine.
I haven’t looked into this because my hardware is too unstable and I can’t afford to raise my bills. What I am wondering is if this has monetary value or if it is a way to keep track of the points. I am guessing that it is the latter because it makes little sense to divert part of your computing power toward acquiring digital currency, though I would say the temptation is there.
Well you’d have to make a lot of curew, since 1 cure coin is worth $0.06 USD: Curecoin price today, CURE live marketcap, chart, and info | CoinMarketCap
Interesting.
For the records, I didn’t mean to refrain people to join this cause, which goal is obviously for the good.
My comment on monetizing was restricted to one of the teams that joined the project, not to the model itself.
That team chose to attract members by giving digital currency in return, and I was surprised that it seemed to be what enabled it to become #1.
On my side, I’m still there, up and running.
1 Like
1 Like
I don’t understand how the points are calculated. I got a work unit which gives more than 7000 points…
Don’t know either, but I think WU’s have different weights. The last ones I received were giving around 1000 each. Maybe they infer people’s processing capability somehow and start giving them adequate WU’s. I remember my first one was huge, and it took a week to process it. Now, with smaller WU’s, it’s taking one or two days.
EDIT: See how smaller this one is?
