Making the cloud smarter
NVIDIA and Microsoft to boost AI cloud computing with new Hyperscale GPU accelerator.
NVIDIA and Microsoft recently unveiled its blueprint to create a new hyperscale GPU accelerator that would enable both companies to drive AI cloud computing.
Known as the HGX-1, the new Hyperscale accelerator is an open-source design released in conjunction with Microsoft’s Project Olympus.
“AI is a new computing model that requires a new architecture,” Jen-Hsun Huang, CEO of NVIDIA said about the company’s collaboration with Microsoft. “The HGX-1 hyperscale GPU accelerator will do for AI cloud computing what the ATX standard did to make PCs pervasive today. It will enable cloud-service providers to easily adopt NVIDIA GPUs to meet surging demand for AI computing.”
The HGX-1 will be powered by eight NVIDIA Tesla P100 GPUs fitted in each chassis, and will utilize an innovative switching design based on NVIDIA’s NVLink interconnect technology and PCIe standard. Once completed, the HGX1 will allow for optimal performance, regardless of the workload.
On paper, the machine will provide up to 100 times faster deep learning performance compared with legacy CPU-based servers, and it is forecasted that it will only require a fifth of the cost to run legacy CPU-based servers in order to conduct AI training and a tenth of the cost for AI inferencing.