High performance computing is the name of the game, while 64-bit is the hook. NVIDIA is readying itself to jump in head first with a number of server vendors to bring on the world’s first 64-bit ARM development systems, all coming our way this year. While the first ARM64 server processors were made for web servers and micro-servers, NVIDIA’s CUDA 6.5 platform brings the heat to HPC-class workloads.
High performance computing systems are used around the world to fuel scientific discovery at locations like Los Alamos National Laboratory. There you’ll find senior scientist Pat McCormick signing in on future NVIDIA CUDA-driven computing with ARM technology as soon as possible.
"We aim to leverage the latest technology advances, both within and beyond the HPC market," said McCormick, "to move science forward in entirely new ways." McCormick continued, "We are working with NVIDIA to explore how we can unite GPU acceleration with novel technologies like ARM to drive new levels of scientific discovery and innovation."
We’ll be seeing the first NVIDIA GPU-accelerated ARM 64-bit development platforms arriving in July of 2014. There you’ll see such systems as the Cirrascale RM1905D, a two-in-one 1U server ready to roll for private cloud consumers with two Tesla K20 GPU accelerators.
There’ll also be an E4 EK003 coming with 3U, dual-motherboard action and two Tesla K20 GPU accelerators as well. Eurotech will be jumping in on the action with a modular Aurora HPC server configuration later this year.