In addition to revealing its plans to license Keplar GPU cores, Nvidia said on Tuesday that its new CUDA 5.5 release candidate brings the power of GPU-accelerated computing to ARM platforms. Now programmers have a robust, easy-to-use suite to develop high-performance computing platforms on both x86 CPU-based and ARM systems.
"Since developers started using CUDA in 2006, successive generations of better, exponentially faster CUDA GPUs have dramatically boosted the performance of applications on x86-based systems," said Ian Buck, general manager of GPU Computing Software at Nvidia. "With support for ARM, the new CUDA release gives developers tremendous flexibility to quickly and easily add GPU acceleration to applications on the broadest range of next-generation HPC platforms."
Nvidia said that thanks to a combination of low-power ARM-based SoCs and CUDA-enabled accelerators, ARM-based systems can now penetrate new markets that require the highest levels of energy-efficient compute performance. That means ARM-based solutions could be used in defense systems, robotics, scientific research and more.
In addition to adding support for ARM architecture, the new toolkit features Hyper-Q support across multiple MPI processes on all Linux systems, MPI Workload Prioritization, new guided performance analysis, and fast cross-compile on x86. This latter feature reduces development time for large applications by enabling developers to compile ARM code on fast x86 processers, and transfer the compiled application to ARM.
CUDA 5.5 also offers a full suite of programming tools, GPU-accelerated math libraries and documentation for both x86- and ARM-based platforms, the company said. GPU-accelerated math libraries include FFT, RNG, BLAS, sparse matrix operations, and nearly 5,000 signal- and image-processing primitives in the NVIDIA Performance Primitives (NPP) library.
Additional information about the new suite can be accessed here on Nvidia's developer website.