Folding@Home set to break 1000TFLOPs (PFLOP)

Stanford ^ | 3-24-2007 | self

[Bogey78O]

http://fah-web.stanford.edu/cgi-bin/main.py?qtype=osstats

Stanford's Folding@Home is set to break into the PFLOP territory. At the current growth (bolstered by the PS3) it should officially break into petFLOPs sometime late Sunday to mid-Monday.

[Bogey78O]

Exciting news for researchers.

[Incorrigible]

Bump

[Bogey78O]

78 more TFLOPs to go till we hit a petaFLOP.

[texas booster]

OS Type...........Current TFLOPS*....Active CPUs... Total CPUs

Windows.....................153..............160471...........1626988

Mac OS X/PowerPC......7...............8801...............95463

Mac OS X/Intel.............9................2880...............7452

Linux............................. 43..............25267............216117

GPU.............................. 43................733...............2228

PLAYSTATION®3.......697..............28441.............32064

Total..............................952............226593..........1980312


Not far from the magic petaFLOP barrier, at 4:40 am CST 3/25/07.

It is amazing that 32,000 PS3s are already crunching away. While the PS3 doesn't solve all of the calculations required to fold a protein, it does run the Implicit Water Molecule Solvents amazingly fast.

[texas booster]

The total teraFLOPS for F@H is now at 972 teraFLOPs.

PS3 makes up 720 TFLOPs with over 29,000 users.

There is a F@H party Sunday night hosted by a gaming site that is asking folks to download F@H and fold for the night.

Hopefully folders will keep folding after the evening is over.

I am amazed that F@H has gained 30,000 new systems to fold in just a week. Thank you Sony. Wow.

Also, shout out to nerdwithamachine for joining the FRee Republic Folding@Home team!

[grey_whiskers]

Details, please?

Can you give me a link to the relevant parts of the Folding@Home website?

(Solvent effects have always been the bane of protein simulations.)

Cheers!

[ThomasThomas]

Sp once they get their proteins folded what do they do with them?

[texas booster]

http://folding.stanford.edu/results.html

All results from the Stanford F@H project are made available to other researchers. So far about 50 papers have been authored by the F@H group or by others linked to them.

[texas booster]

http://www.youtube.com/watch?v=VoMHnKMD_Aw&mode=related&search=

Sure its a puff piece on the Sony, but Dr Pande makes an appearance!

The wiggles in the protein are supposed to be Brownian motion of the mater molecules. I'll keep searching to find the discussion of solvent effects and how the PS# assists.

[texas booster]

F@H uses the TINKER and GROMACS core to fold proteins, and there are other newer cores available to use.

From the High Performance FAQ at Stanford U:

Why push for new technologies, such as GPUs or the PS3, rather than try to recruit more CPU clients?

In order to tackle many of the problems of interest (especially related to protein misfolding and aggregation, such as in Alzheimer's Disease), we need to not just have lots of computers participating, but we need results returned more quickly so that we can simulate trajectories of sufficient length. Right now, we achieve this by running for many months to years (indeed, our first Alzheimer's Disease simulations ran for almost two years straight). That's where the new clients come in. They give us considerably longer trajectories in the same wall clock time, allowing us to turn what used to take years to simulate even on FAH, to a few weeks to months.

What types of calculations do the new clients speed?

There are primarily two types of calculations we perform, which differ by how we simulate water. The GPU and PS3 clients greatly speed "implicit solvation" simulations, in which water is handled mathematically in a continuum fashion (see the Wikipedia article on implicit solvation for more information). Our SMP client will significantly speed "explicit solvent" simulations, where water atoms are handled atom by atom, in an explicit fashion, just like any other atom in our system. Currently, the GPU & PS3 only significantly speed implicit solvation and the SMP client only speeds explicit solvent, so each has its limits, but together they work to give FAH considerably more computational power than ever before.

Links cited above:

Wiki

[texas booster]

Also, you could ping Dan Ensign over at the F@H Forum.

F@H is one of the very few DC projects where you can ping the actual developers, and even the PhD in charge, Vijay Pande, and get answers.

Check out this link. If you see something interesting, ping Dan. He is terrific in replying quickly.

http://forum.folding-community.org/ftopic14500.html

[Bogey78O]

12 more teraFLOPs left till we hit the next milestone.