PC News Zone » GPU Computing

KQED Warns Of Supercomputing â€˜Energy Wallâ€™ â€“ NVIDIA Helps Scale It With GPUs

PC News Zone — Sat, 02 Jul 2011 09:00:47 +0000

Lauren Sommer wrote a great blog over the weekend on KQED about how supercomputers have hit the “energy wall” – a decidedly real supercomputing problem that NVIDIA’s GPU technology can help to overcome. This is what 1,000 homes looks like. The blog post mentions the Hopper supercomputer, located at the Lawrence Berkeley National Lab (LBNL). The system consumes 3 megawatts of electricity (enough to power 2,000-3,000 homes a year) and has performance of 1 petaflop per second (equivalent to about 68,000 laptops). It’s hard to imagine these numbers scaling to exascale systems – the “energy wall” here would just be too high to reasonably surmount. In fact, I just got back from the International Supercomputing Conference in Hamburg, where the running joke was that power companies would soon be giving supercomputers away for free if you sign up for a five-year power contract with them. Here at NVIDIA, we’ve been working on a solution to the supercomputing power crisis for several years. Supercomputers can use NVIDIA Tesla GPUs to dramatically accelerate supercomputing applications. Like a turbocharger on your car, GPUs kick in to boost your standard Intel or AMD CPUs when you need the extra oomph. Using GPUs is a much more energy efficient way of supercomputing. You choose the right processor to the do the right job. When I edit pictures of my kids, for example, my computer’s sequential Intel or AMD x86 CPU is used to access the hard disk, retrieve the file, and open it. Once the picture is open, and I want to do red-eye reduction or remove the blur, the GPU kicks into gear to accelerate the job. Three of the Top Five supercomputers in the world are accelerated by NVIDIA Tesla GPUs. One of these is the Tsubame 2.0 system at the Tokyo Institute of Technology. Like the Hopper system at LBNL, it delivers 1 petaflop per second of performance. But thanks to its GPUs, it consumes less than half the power of the Hopper system. To be exact, Tsubame achieves 1.19 Petaflop/sec and sips a “mere” 1.4 megawatts of electricity. Half the power for the same performance is a big leap forward. But we have a long road ahead, especially as we move towards exascale supercomputers that will be 1,000 times more powerful than the current petaflop supers. Otherwise, the power companies will indeed start giving away supercomputers away for free!

Blog: Microsoft Going All-in on GPU Computing

pcnz_admin — Wed, 15 Jun 2011 17:50:36 +0000

Great newsâ€¦â€¦Microsoft today made an announcement that will accelerate the adoption of GPU computing (that is, the use of GPUs as a companion processor to CPUs). The software maker is working on a new programming language extension, called C++ AMP, with a focus on accelerating applications with GPUs.

With Microsoft now embracing GPUs in their future higher level language and OS roadmap, it makes the decision to go with GPU computing even easier for those programmers still on the fence.

Its intent with C++ AMP is to expose C++ language capabilities to millions of Windows developers with the goal of enabling them to take advantage of GPUs. It promises to give millions of C++ developers the option of using Microsoft Visual Studio-based development tools to accelerate applications using the parallel processing power of GPUs. CUDA C and CUDA C++ will continue to be the preferred platform for Linux apps or demanding HPC (high performance computing) applications that need to maximize performance.

In the Spring 2007, there was just one language (CUDA C) supporting NVIDIA GPUs. Fast forward to today and our customers now have a much wider selection of languages and APIs for GPU computing â€“ CUDA C, CUDA C++, CUDA Fortran, OpenCL, DirectCompute and in the future Microsoft C++ AMP. There are even Java and Python wrappers, as well as.NET integration, available that sit on top of CUDA C or CUDA C++.

If you are a Windows C++ developer looking at GPU Computing for the first time, there is no need to wait. Visual C++ developers today use our high performance CUDA C++ with the Thrust C++ template library to easily accelerate applications by parallelizing as little as 1 to 5 percent of their application code and mapping it to NVIDIA GPUs. CUDA C++ comes with a rich eco-system of profilers, debuggers, and libraries like cuFFT, cuBLAS, LAPACK, cuSPARSE, cuRAND, etc. NVIDIAâ€™s Parallel Nsightâ„¢ for Visual Studio 2010 provides these Windows developers a familiar development environment, combined with excellent GPU profiling and debugging tools.

The take away from Microsoftâ€™s announcement today is that the GPU computing space has reached maturity, with the company that produces the worldâ€™s most widely used commercial C++ developer tools â€“ Microsoft â€” completely embracing GPU computing in their core tools. Rest assured, NVIDIA continues to work closely with Microsoft to help make C++ AMP a success, and we will continue to deliver the best GPU developer tools and training.

Stay tuned for more details.