NVIDIA Tesla GPUs now shipping with Dell 'personal supercomputers'
By Joseph L. Flatley 
posted May 6th 2009 3:18PM
posted May 6th 2009 3:18PM
Been itching to get your hands on a personal supercomputer, as NVIDIA's ad wizards put it? The company has just announced that its CUDA-based Tesla C1060 GPU is now available in Dell's Precision R5400, T5500 and T7500 workstations. And just to put things into perspective, NVIDIA points out that a Dell workstation rockin' a single Tesla C1060 has enough going on under the hood to power the control system for the European Extremely Large Telescope project ("the world's largest," apparently). According to one of the developers, Jeff Meisel at National Instruments, a workstation "equipped with a single Tesla C1060 can achieve near real-time control of the mirror simulation and controller, which before wouldn't be possible in a single machine without the computational density offered by GPUs." Wild, huh? If you're curious about the workout that Tesla GPUs are getting on a wide range of projects, from Bio-Informatics to Computational Chemistry to Molecular Dynamics and more -- or if you're merely a glutton for long-winded PR -- check out the good stuff after the break.
Personal SuperComputing With DELL: NVIDIA Tesla GPUs Now Shipping In DELL Workstations
NVIDIA Tesla GPU Computing Solutions Become Mainstream as Number of GPU Optimized Applications Soars
For further information, contact:
Andrew Humber
NVIDIA Corporation
(408) 416 7943
ahumber@nvidia.com
FOR IMMEDIATE RELEASE:
SANTA CLARA, CA -MAY 6, 2009- NVIDIA Corporation, inventor of the GPU, today announced that the Tesla™ C1060 GPU Computing processor, based on the massively parallel CUDA™ architecture, is now available in Dell Precision R5400, T5500 and T7500 workstations.
"The Dell Precision R5400, T7500 and T5500 together with the Tesla GPU computing processors is putting the power of supercomputing on the desktop," said Greg Weir, senior manager, Dell Product Group. "We have seen early praise for the efforts of both Dell and NVIDIA to bring an economical high-performance computing solution to the most demanding customers."
"National Instruments is developing the control system for the European Extremely Large Telescope project, which upon completion will be the world's largest. To tackle this computational challenge, we developed a CUDA interface with LabVIEW to simulate and control the M1 mirror consisting of 984 individual segments," said Jeff Meisel product manager for LabVIEW at National Instruments. "A Dell workstation equipped with a single Tesla C1060, can achieve near real-time control of the mirror simulation and controller, which before wouldn't be possible in a single machine without the computational density offered by GPUs."
Another community sure to benefit from the mass market availability of this technology is the computational researcher. Based in the world's leading research schools such as Harvard, Cambridge or Tokyo Institute of Technology, these researchers fight for time on a shared supercomputing resource that consumes hundreds of kilowatts of power and costs millions of dollars to build and maintain. Dell Precision Workstations enabled with Tesla GPUs give each of these researchers their own "personal supercomputer" - the equivalent computing power of a cluster, at 1/100th of the price.
CUDA applications actively in use today by these researchers and organizations include:
Oil and gas
* Acceleware: Kirchoff Time Migration library
* ffA: 3D Seismic processing software
* Headwave: Prestack data processing
* Mercury Computer systems: 3D data visualization
* SeismicCity: 3D seismic imaging for prestack depth migration
* SMT: Kingdom – Seismic Processing
Computational Chemistry and Molecular Dynamics:
* GROMACS molecular dynamics
* HOOMD molecular dynamics
* NAMD molecular dynamics
* VMD visualization of molecular dynamics
Bio-Informatics and Life Sciences:
* GPU HMMER: CUDA version of HMMER
* LISSOM: Human neocortex modeling
* MUMmerGPU: High-throughput DNA sequencing
Financial Computing and Options Pricing:
* Aqumin: 3D Visualization of market data
* Exegy: Risk Analysis
* Hanweck: options pricing
* SciComp: derivatives pricing
Mathematical Computing
* Jacket CUDA plugin for MATLAB from Accelereyes
* LabVIEW from National Instruments
GeoSciences:
* Tsunami simulation – Tokyo Institute of Technology
* Weather Research and Forecast (WRF) model
* Geographical Information Systems - Manifold
Medical Imaging, CT, MRI:
* AxeRecon CT reconstruction library from Acceleware
* SnapCT tomographic reconstruction software from Digisens
Electrodynamics and Electromagnetics
* CST Microware Studio
* FDTD solver from Acceleware
Electronic Design Automation
* ADS SPICE simulator from Agilent EESof
* OmegaSim GX SPICE simulator from Nascentric
* Sentaraus TCAD from Synopsys
For more information on the Dell Precision Workstation line, visit www.dell.com. For more information on NVIDIA Tesla products, visit www.nvidia.com/object/tesla_computing_solutions and for more information on applications written for the CUDA architecture, visit www.nvidia.com/cuda.
But can it play: Monkey Island 2: LeChuck's Revenge?
But .... Will it blend ? (hope no one made that before on this post)
Engadget doesn't think it's newsworthy, but you can now configure 192 Gb of RAM in the Precision T7500 if you go to large business on their website. It will only cost you a mere $229,000 extra, lol.
I can't stand LabVIEW.
On a side note, the 42 meter diameter European Extremely Large Telescope project is going to be INSANE! The current largest telescope is the world is the twin Keck telescopes in Hawaii, which have a diameter of 10 meters. Using some 6th grade math, area = Pi r^2, we get
Keck 10m ----> 3.14 * 5^2 = 79 square meters
European ELT 42m ---> 3.14 * 21^2 = 1385 square meters
So the European ELT has a 17X larger light collection area than the Keck! Thats crazy! this thing will be able to actually study the atmospheres of extrasolar planets looking for life
Can it play..... God?
How much would one of these cost in, say, Internets?
1
Can it play Half-Life 3?
i bet it cant even play pong!
I saw the demo for this project at the Supercomputing conference in Austin last year. Very impressive. They put this system up against a quad core from Dell and for the this type of application, the Tesla smoked it.
Sounds like good for folding :D
what about making breakfast?
Just to add some more details on the LabVIEW application mentioned above which is in the proof-of-concept stages for ESO, it's in fact a really difficult computational problem when you consider the physical design of the E-ELT is years from being completed. What that means is both the simulation of the telescope and how all the mirror segments interact (which doesn't yet exist, since it's not built) and the control have to be performed as if it were happening in "real-time". We've tried a number of approaches to tackling the problem, and as the article mentions one configuration that shows a lot of promise is a Dell T7400 with a Tesla C1060, where we separate the problem and do the simulation on the T7400 and the real-time control on the C1060. This approach is unique in that it's the only one we've looked at that can be implemented in a single machine (whereas others require multiple machines in more of a traditional HPC configuration). We're still in the prototype stages but our Tesla C1060 GPU benchmarks have been impressive, making it a very viable option moving forward.
Jeff Meisel, LabVIEW Product Manager, National Instruments