Pure Storage partners with Nvidia to deploy AI-ready storage infrastructure

Pure Storage partners with Nvidia to deploy AI-ready storage infrastructure
28 Mar, 2018

Flash storage company Pure Storage has partnered with graphic processing unit producer Nvidia to come up with a new artificial intelligence-ready infrastructure solution to boost productivity for data scientists.

The product, AIRI, or AI-ready infrastructure, comes with Flash Blade storage solutions. Nvidia DGX-1 servers and workstations, which specialise in general-purpose computing on graphics processing units to accelerate deep learning applications, will power AIRI.

The product is capable of storing 50 racks of computing and storage server, but will occupy just half the storage space. “It is basically giving the performance of an entire data science center...to one data scientist team in something less than a full rack in size and scale,” Matt Burr of Pure Storage was quoted as saying by ZDNet.

“No organisation can take the money it takes to do this today. To give every one of their data scientists this kind of power. It is a really meaningful jump,” he said, adding that AIRI was built for enterprises or customers who were looking to get into AI.

Explaining further, Burr said that while the product might have anticipated growing requirements of hyperscale computer users and research academics, enterprises these days are focused on “how do I buy things that are relatively generic, such that I can apply across a large topology of infrastructure space and yield a pretty solid result?”

“We reached a point where we said if these really smart companies...were looking for simplified infrastructure to onramp to artificial intelligence and analytics, AIRI is probably the answer," Burr was quoted as saying.

The combination of technology in AIRI allows the system to deliver four petaFLOPS of performance with NVIDIA® Tesla V100 GPUs. “These systems are interconnected with Arista 100GbE switches, supporting GPUDirect RDMA for maximum distributed training performance,” said Jim McHugh, vice-president and general manager of DEEP LEARNING SYSTEMS at Nvidia.

“AIRI is also supported by the NVIDIA GPU Cloud deep learning stack and Pure Storage AIRI Scaling Toolkit, enabling data scientists to jumpstart their AI initiatives in hours, not weeks or months,” he added.

Last week, Vishal Dhupar, who heads Nvidia’s India subsidiary, had told TechCircle that the company was looking to work with startups and big firms to build an AI ecosystem in the country.