Deep Learning with IBM OpenPower and NVIDIA Tesla P100 (Pascal) GPU
Deep learning, or the use of multi-layer neural networks, has revolutionized speech recognition, natural language processing, and computer vision, and continues to revolutionize IT due to availability of rich data sets, new methods for accelerating neural network training and extremely fast hardware with GPU accelerators.
POWER8 is ideal for deep learning, big data, and machine learning due to its high performance, large caches, 2x-3x higher memory bandwidth, very high I/O bandwidth, and of course, tight integration with GPU accelerators. POWER8’s parallel, multi-threaded architecture with high memory and I/O bandwidth is particularly well adapted to ensure that GPUs are used to their fullest potential.
The session will present latest IBM OpenPower server with POWER8 CPUs connected by the high-speed NVLink interface directly to NVIDIA Tesla P100 (Pascal) GPU accelerators and related performance for Deep Learning frameworks.