Yesterday, we posted photos from our exhibit. Today was the last day of SC13, and we want to tip our hat to the wonderful partners that magnified our SC13 experience. Creative Consultants, Mellanox, and Allinea Creative Consultants ran an ArrayFire demo across several nodes using Mellanox interconnect. The demo was a multi-node, multi-GPU lattice boltzmann simulation. Allinea also showcased their debugging and profiling tools on the same ArrayFire based code. AMD ArrayFire OpenCL demos were showcased in the AMD exhibit. It was great to see momentum from AMD at SC13 carried over from the previous week’s APU13 conference. Microway In the photo below, you can see ArrayFire running on Microway’s WhisperStation. Microway had prime real estate at the conference and surely every …
How much speedup can you get with CUDA or OpenCL?
Everyday developers ask us to predict how much speedup they can get with CUDA or OpenCL. Rather than gaze mysteriously into a crystal ball, we ask the developers questions to explore pertinent acceleration factors. Note, we’ll use the term accelerator to include GPUs, Xeon Phi coprocessor, APUs, FPGAs, and any other CUDA or OpenCL device. The principles we discuss below are equally applicable to all of these accelerators. The following are some of the important factors that must be considered when estimating the potential for accelerated speedups: Hardware: The more advanced the accelerator hardware, the more the speedup you get (e.g. the NVIDIA Kepler K20 outperforms the previous NVIDIA Fermi C2090 generation). Data Sizes: In general, accelerators will outperform CPUs to …