Hardware Canucks

Hardware Canucks (http://www.hardwarecanucks.com/forum/)
-   Press Releases & Tech News (http://www.hardwarecanucks.com/forum/press-releases-tech-news/)
-   -   All you folders can hang up your hats.... (http://www.hardwarecanucks.com/forum/press-releases-tech-news/414-all-you-folders-can-hang-up-your-hats.html)

Babrbarossa February 22, 2007 08:53 AM

All you folders can hang up your hats....
I'm not sure how many fps you'd get on COH with this baby, but this molecular dynamics simulation rig could probably fold faster than Rita McNeale's maid on laundry day.

"What's the world's fastest supercomputer used for?

The world's fastest supercomputer will probably never be known as the world's fastest supercomputer. RIKEN's MDGrape-3 is the first machine to break the petaflop barrier -- that's 1 quadrillion calculations (floating-point operations, to be specific) per second -- and it's three times faster than the currently ranked fastest computer in the world, IBM's BlueGene/L. But MDGrape-3 is so specialized that it can't run the software used to officially rank computing speed. What it can do is determine the effect of any chemical compound on one of the most intricate systems in the human body in a couple of seconds.

MDGrape-3 is designed for pharmaceutical research, specifically molecular dynamics simulation. In developing drugs, pharmaceutical companies have to analyze thousands on thousands of chemical compounds to find out how they'll affect the protein-bonding structures in the human body. Protein structures called enzymes are the building blocks that do all of the work within a cell, and the way these proteins bond with any drug compound introduced into the human body determines the body's response to that drug. MDGrape-3 produces simulations of these molecular interactions. What takes most computers hours or days to analyze takes MDGrape-3 a few seconds. This functionality is invaluable in drug research, and it could drastically cut the research time involved in the development of new cures. A subsidiary of pharmaceutical giant Merck has already booked time on the machine.
Structurally speaking, MDGrape-3 is a parallel computing system consisting of two main sections: a primary server unit and a specialized-engines unit. The latter component is a cluster of 201 engines running proprietary chips developed by Riken specifically for MDGrape-3. It's this huge set of engines, running 24 MDGrape-3 chips each, that does the heavy protein-analysis lifting. Each chip has a maximum processing speed of 230 gigaflops (one billion operations per second). The primary server unit manages the engine cluster. This parallel server setup runs two different types of processors: 65 servers run dual-core Intel 5000-series Xeon processors, 256 per server; and 37 servers run 3.3-GHz Intel Xeon processors, each with 2 MB of level 1 cache, at 74 processors per server. This hardware structure enables the 1-petaflop speed, which is the machine's theoretical maximum for certain processes.
MDGrape-3 took $9 million and about four years to build. And it's actually very efficient -- a total cost of $9 million breaks down to about $15 per gigaflop. The slower BlueGene/L cost about $140 per gigaflop to build. BlueGene/L, which tops out at a theoretical 360 teraflops (trillion calculations per second), is also a biotechnology-specific machine. The advances in speed marked by these two supercomputers is indicative of a general trend in technology toward biologically-slanted systems. Some say the trend really started with the successful mapping of the human genome in 2000. Regardless of what spurred the current biotechnology race, most experts agree that the logical end of the surge is a state of DNA-based medicine. In several decades, we could make an appointment with our doctor for a quick DNA analysis to find out what diseases we're at risk for and pop a single, gene-targeting pill that eliminates all of those foreseeable risks. "

from HowStuffWorks

Supergrover February 22, 2007 03:25 PM

Well I think theres what a million folders @ least? I dont think we will stop just yet.

I believe F@H started because supercomputers just couldn't do the job as compared to millions of computers working on the same protein. Sure supercomputers have come a long way but so has the PC.

Babrbarossa February 22, 2007 04:07 PM

I was completely joking

Supergrover February 22, 2007 07:01 PM


Originally Posted by Babrbarossa (Post 2130)
I was completely joking

Doh My Bad:ph34r:
I was just stating that this will never equate to the shear combined power of the F@H client. Think single cores vers Multi core, not quite right but good enough for this statement.:bleh:

sswilson February 23, 2007 08:13 AM

I've just started running the gpu client, and boy is it fast!!!!!

It got me thinking about how the integration of ati's chipsets & gpu's with AMD's procs could really boost performance.

The quoted # I've seen in articles has been about 70X more processing power for the gpu doing this kind of # crunching.

Just imagine what kind of improvements in power you'd get if the chipset was designed to offload appropriate non-graphical processes to the graphics card.

Babrbarossa February 23, 2007 08:58 AM

I hear that it's a real stress on the system- is it driving your temps really high?

sswilson February 23, 2007 09:49 AM

For the graphics card it's no worse than playing heavy 3D. Strangely enough it stresses the CPU more than any other stress test I've run......

Supergrover February 23, 2007 02:19 PM


Originally Posted by sswilson (Post 2167)
For the graphics card it's no worse than playing heavy 3D. Strangely enough it stresses the CPU more than any other stress test I've run......

Damn straight F@H is better stabilty program than orthoes, I have been 24 prime but still blew workunits, there is like zero room for error with gromacs, tinkers you dont need to be stable, you got to be close but tinkers hardly stress the system.

Whats you ppd on that gfx wilson?

sswilson February 23, 2007 03:18 PM


Is that something I can find in the log file?

As far as stress testing goes.... it does a fine job of it, but it's not consistant.

Prime is a good benchmark because it's the same every time you run it. FAH depends on what WU you get.

Supergrover February 23, 2007 03:40 PM

ppd=Points Per Day

Wait for the large dgromacs cores those are the ones that stress your system.

If your not throwing those type of WU ( WU=workunit ) then your rock solid stable.

To check what wu you have go Here this gives you the points for the wu and what servers are running what wu.
Go Here to find out what servers are down.

I think I might do up a F@H F.A.Q. since it seams we have a few folders here.

How would you guys feel about starting a HCF team?

All times are GMT -7. The time now is 06:18 AM.