At this week's SC08 conference in Austin, Texas, NVIDIA announced several breakthroughs with their Tesla supercomputer.
NVIDIA Tesla Makes Personal Supercomputing a Reality
Today, scientific research is carried out on supercomputing clusters, a shared resource that consumes hundreds of kilowatts of power and costs millions of dollars to build and maintain. As a result, researchers must fight for time on these resources, slowing their work and delaying results. NVIDIA and its worldwide partners have releaesd the GPU-based Tesla Personal Supercomputer, which delivers the equivalent computing power of a cluster, at 1/100th of the price and in a form factor of a standard desktop workstation.
"We've all heard 'desktop supercomputer' claims in the past, but this time it's for real," said Burton Smith, Microsoft Technical Fellow. "NVIDIA and its partners will be delivering outstanding performance and broad applicability to the mainstream marketplace. Heterogeneous computing, where GPUs work in tandem with CPUs, is what makes such a breakthrough possible."
Priced like a conventional PC workstation, yet delivering 250 times the processing power, researchers now have the horsepower to perform complex, data-intensive computations right at their desk, processing more data faster and cutting time to discovery.
"GPUs have evolved to the point where many real world applications are easily implemented on them and run significantly faster than on multi-core systems," said Jack Dongarra, director of the Innovative Computing Laboratory at the University of Tennessee and author of LINPACK. "Future computing architectures will be hybrid systems with parallel-core GPUs working in tandem with multi-core CPUs."
Leading institutions including MIT, the Max Planck Institute, University of Illinois at Urbana-Champaign, Cambridge University, and others are already advancing their research using GPU-based personal supercomputers. "GPU based systems enable us to run life science codes in minutes rather than the hours it took earlier. This exceptional speedup has the ability to accelerate the discovery of potentially life-saving anti-cancer drugs," said Jack Collins, manager of scientific computing and program development at the Advanced Biomedical Computing Center in Frederick Md., operated by SAIC-Frederick, Inc.
At the core of the GPU-based Tesla Personal Supercomputer is the Tesla C1060 GPU Computing Processor which is based on the NVIDIA CUDA parallel computing architecture. CUDA enables developers and researchers to harness the massively parallel computational power of Tesla through industry standard C.
"Dell has led the workstation category for almost a decade and GPU computing represents a massive leap forward in performance that will bring supercomputer power to the masses," said Antonio Julio, director, Dell Product Group. "The Dell Precision R5400 and T7400 will allow the scientific community to harness the capabilities of the NVIDIA Tesla C1060 GPU with up to two teraflops of computational power."
As well as Dell, GPU-based Tesla Personal Supercomputers are available today from the following leading HPC OEMs, Systems Builders and Resellers: AMAX (U.S.), Armari (U.K.), AsU.S. (WW), Azken Muga (ES), Boxx (U.S.), CAD2 (U.K.), CADnetwork (DE), Carri (FR), Colfax (U.S.), Comptronic (DE), Concordia (IT), Connoisseur (IN), Dell (WW), Dospara (JP), E-Quattro (IT), JRTI (U.S.), Lenovo (WW), Littlebit (CH), Meijin (RU), Microway (U.S.), Sprinx (CZ), Sysgen (DE), Transtec (DE),Tycrid (U.S.), Unitcom (JP), U.S.tar (U.K.R),Viglen (U.K.) and Western Scientific (U.S.).
To learn more about the industry-changing applications benefitting from NVIDIA GPU Computing technology, visit www.nvidia.com/cuda and for more information on the GPU-based NVIDIA Tesla Personal Supercomputer, visit www.nvidia.com/personal_supercomputing.
NVIDIA Tesla Gives Bull Customers a Revolutionary Performance Boost
Bull, a supplier of high performance computing (HPC) technologies, is partnering with NVIDIA to provide the Tesla S1070 GPU Computing System as the accelerator option for their HPC solutions. With this announcement, HPC customers in Europe can now get integrated solutions and services from Bull that combine Tesla GPU Computing solutions with Bull's range of other products for HPC.
"NVIDIA's Tesla parallel GPU technology is truly disruptive, delivering performance increases that are transforming industries," said Fabio Gallo, VP and director of HPC Solutions at Bull. "The Tesla S1070 is a world-class solution for HPC users wanting to handle large datasets and solve complex problems fast."
In addition, Bull has assembled a team of technical experts to work with customers and optimize their Tesla based solutions to deliver best performance for their given workload. This team is also able to provide training and other services as an integral part of a full service solution.
GENCI - the French national High-Performance Computing Agency -- together with CEA, the French Atomic Energy Authority -- is one of the many organizations utilizing the massive parallel processing power of NVIDIA Tesla in a large scale supercomputing project. Built and installed by Bull, the NovaScale supercomputer takes the available power at the scientific computing complex in Bruyeres le Chatel (in the Ile de France region) to more than 300 Teraflops, making it one of the most powerful hybrid supercomputer in Europe.
The Tesla S1070 1U system is based on the NVIDIA CUDA parallel architecture. This architecture is accessible through an industry standard C language programming environment that allows developers and researchers to tap into the parallel architecture of the GPU more quickly and easily than any other solution shipping today.
For more information on NVIDIA Tesla S1070, visit: www.nvidia.com/object/tesla_s1070.
NVIDIA Tesla Turbocharges High-Performance Computing Industry with HP ProLiant Servers
NVIDIA's Tesla S1070 Computing System is now being offered in the highly successful range of HP ProLiant servers. Several Tesla GPU based systems are already installed and are transforming a wide range of industries.
Acceleware, a company that specializes in enabling software vendors to leverage parallel processing architectures, has been working with Kodak to deliver a solution of clustered Tesla GPU servers for accelerating their large engineering simulations.
"Acceleware's cluster solution on NVIDIA Tesla GPU server-accelerated HP hardware enables Kodak to develop market-leading, high-performance image sensor products for consumer, professional, and advanced applications," said Herb Erhardt, Manager of the CMOS Sensors business unit at Kodak. "We have already seen significant productivity improvements that amount to greater than 10 times."
"Customers can respond rapidly to changing markets and maintain a competitive advantage by dramatically increasing the performance of their systems," said Ed Turkel, product marketing manager, Scalable Computing and Infrastructure organization at HP. "Through participation in the HP Accelerator program, NVIDIA Tesla GPUs on HP ProLiant servers deliver unparalleled performance boosts and, in some cases, enables computations not possible with previous generation technologies."
Tesla GPUs are available through HP's innovative HPC Accelerator program -- for more information, visit: www.hp.com/go/accelerators. For more information on NVIDIA Tesla S1070, visit: www.nvidia.com/object/tesla_s1070.
NVIDIA and Cray to Deliver Tesla-Enabled Cray CX1 Deskside Supercomputer
The NVIDIA Tesla C1060 GPU Computing processors in the new Cray CX1 line of supercomputers are now available. With "ease-of-everything" features and the ability to fit into a standard office setting, the Cray CX1 product reflects NVIDIA and Cray's goal to drive high productivity computing solutions into a broad array of markets including financial services, oil and gas, life sciences, government and academic.
"NVIDIA GPU computing technology can powerfully accelerate computational capability on important classes of applications. The Cray CX1 system is a flexible platform for individual scientists, workgroups and departments, and coupled with NVIDA's Tesla computing processors and Quadro GPUs, the Cray CX1 delivers an industry-defining supercomputer", said Ian Miller, senior vice president of sales and marketing at Cray.
Each Tesla processor has hundreds of processor cores that deliver nearly one teraflop of peak computing performance. The Cray CX1 delivers up to four teraflops of performance, right at the deskside, when configured with four Tesla processors.
"Many organizations today are investigating ways to augment the computational resources of their cluster and increase productivity," said Andy Keane, GM of the GPU Computing business at NVIDIA. "The Cray CX1 combined with NVIDIA Tesla GPUs makes massive compute power accessible to scientists and engineers, transforming their workflow and enabling them to get results fast, dramatically increasing the pace of discovery."
Using GPUs, researchers have reported up to 100x speed-up on applications in fields such as molecular dynamics, quantum chemistry, finite element analysis, particle simulation, and electronic design automation. The highly parallel architecture of the GPU has been made accessible for these industries through the NVIDIA CUDA architecture. With NVIDIA's award-winning C-compiler and software development kit (SDK) for developing parallel computing applications on GPUs, developers can exploit the GPU's parallel computing architecture and automatically distribute computing work across tens of thousands of threads and hundreds of processor cores.
NVIDIA and NEC Collaborate to Deliver GPU Computing Solutions to HPC Market
NVIDIA has begun a close collaboration with NEC to integrate NVIDIA Tesla GPUs into its systems for the high performance computing (HPC) industry. NEC brings extensive experience as a major solution provider in the HPC market to the partnership which will see powerful, parallel GPU Computing solutions deployed into a range of industries.
"This partnership is an important step for NEC to promote a total solution business including heterogeneous or hybrid computing technologies," said Fumihiko Hisamitsu, GM of NEC's High performance computing marketing promotion division. "NEC will offer unique HPC solutions to our joint customers and will work with NVIDIA globally to specify Tesla solutions for customers designing high performance computing clusters."
The first customer to leverage this collaboration is the Global Scientific Information and Computing Center (GSIC) at Tokyo Institute of Technology, whose TSUBAME supercomputer has lead the supercomputing scene both in Japan and globally for the past 2.5 years. TSUBAME was upgraded in Oct 2008 with 680 NVIDIA Tesla GPUs, unprecedentedly while the system remained in operation. The newly upgraded system recorded 77.48 Teraflops in Linpack, ranking it high in the global Top 500 supercomputer listing.
NVIDIA Tesla GPU Computing processors are revolutionizing industries such as oil and gas, finance, medical and life science. In many cases, processing tasks that are simply not possible on CPU-based clusters and workstations are being enabled by the Tesla 10-series products. Each Tesla processor has 240 cores, providing 1 Teraflop of processing power, along with 4GB of onboard memory. The new NVIDIA Tesla S1070 1U system features four Tesla 10-series processors -- for a total of 960 cores and 4 Teraflops of processing power.