Wondering if ARM
-based chips are powerful enough for the real world? Here's your answer. NVIDIA
has just unveiled a new prototype system with Tegra ARM CPUs and CUDA GPUs, and that system just so happens to be a supercomputer. The company has announced that the Barcelona Supercomputing Center (BSC) is developing a new hybrid supercomputer that, for the first time, uses energy-efficient, low-power NVIDIA Tegra ARM CPUs, together with high-performance NVIDIA CUDA GPUs.
BSC is planning to develop the first large scale system based on this technology, with a near term goal of demonstrating two to five times improvement in energy efficiency compared with today's most efficient systems. BSC's ultimate research goal is to deliver exascale-level performance while using 15 to 30 times less power than current supercomputer architectures. This so-called EU Mont-Blanc Project will explore next-generation HPC architectures and develop a portfolio of exascale applications that run efficiently on these kinds of energy-efficient, embedded mobile technologies.
"In most current systems, CPUs alone consume the lion's share of the energy, often 40 percent or more," said Alex Ramirez, leader of the Mont-Blanc Project. "By comparison, the Mont-Blanc architecture will rely on energy-efficient compute accelerators and ARM processors used in embedded and mobile devices to achieve a four- to 10-times increase in energy-efficiency by 2014." To support growing demand for similar ARM-based initiatives around the world, NVIDIA also announced plans to develop a new hardware and software development kit. The kit, with hardware developed by SECO, will feature a quad-core NVIDIA Tegra 3 ARM CPU accelerated by a discrete NVIDIA GPU. It is expected to be available in the first half of 2012, and will be supported by the NVIDIA CUDA parallel programming toolkit.
Good enough for a supercomputer = good enough for you?