Nvidia launches cloud AI partner programme

Nvidia website under a magnifying glass
(Image credit: Shutterstock)

Nvidia hopes to meet the rapidly growing demands for AI cloud computing by launching a partner programme with some of the world's largest original design manufacturers (ODMs).

The US technology giant, known for its high end graphical processing units, has teamed up with leading manufacturers Foxconn, Inventec, Quanta and Wistron, through its HGX Partner Program.

The scheme will grant ODMs early access to new developments of Nvidia's HGX architecture, the same technology built into data centres used by Microsoft's open source hardware initiative Project Olympus, and Facebook's Big Basin AI platform.

Manufacturers will also receive direct support from Nvidia engineers, which the company hopes will help reduce the time between initial technology design and the launch of new products.

"Accelerated computing is evolving rapidly - in just one year we tripled the deep learning performance in our Tesla GPUs - and this is having a significant impact on the way systems are designed," said Ian Buck, general manager of accelerated computing at Nvidia. "Through our HGX partner programme, device makers can ensure they're offering the latest AI technologies to the growing community of cloud computing providers."

Nvidia's HGX architecture was unveiled in March at the Open Compute Summit in partnership with Microsoft as a data centre-focused alternative to the ATX standard. The new HGX-1 standard is designed to allow GPUs and CPUs to connect in whatever setup is required for the workload.

The programme represents Nvidia's latest push into the field of AI following the launch of its Tesla V100 GPUs in May; a new range of chips built on its Volta architecture and specifically designed to power AI applications.

The HGX architecture packs eight of these GPUs into a single modular chassis, capable of being added to existing data centre racks.

Specifically for AI cloud computing, the architecture is built to host Nvidia's GPU Cloud platform, which provides a catalogue of deep learning suites, including Caffe2, Cognitive Toolkit, MXNet and TensorFlow.

"Working more closely with Nvidia will help us infuse a new level of innovation into data center infrastructure worldwide," said Evan Chien, head of IEC China operations at Inventec. "Through our close collaboration, we will be able to more effectively address the compute-intensive AI needs of companies managing hyperscale cloud environments."

Dale Walker

Dale Walker is the Managing Editor of ITPro, and its sibling sites CloudPro and ChannelPro. Dale has a keen interest in IT regulations, data protection, and cyber security. He spent a number of years reporting for ITPro from numerous domestic and international events, including IBM, Red Hat, Google, and has been a regular reporter for Microsoft's various yearly showcases, including Ignite.