Graphics giant Nvidia introduces next-gen chips for networking, storage and security amidst the new age of artificial intelligence (AI). It announced major schemes for data centres, edge AI, collaboration tools and healthcare.

In the GPU Technology Conference, Nvidia CEO Jensen Huang said, “AI requires a whole reinvention of computing – full-stack rethinking – from chips, to systems, algorithms, tools and the ecosystem. Nvidia is a full-stack computing company. We love working on extremely hard computing problems that have a great impact on the world -- this is right in our wheelhouse. To run modern data centres, a new kind of processor is needed. We call it the data processing unit (DPU). The DPU consists of accelerators for networking, storage, security and programmable Arm CPUs to offload the hypervisor.”

 Nvidia’s new ‘BlueField 2’ DPU is made with powerful arm cores and includes a programmable processor, with acceleration engines for at-line-speed processing for networking, storage and security.

Mellanox is the latest acquisition by Nvidia. It closed in April. Nvidia also announced DOCA, which is a programmable data-center-infrastructure-on-a-chip architecture.

Huang said, “DOCA SDKs let developers write infrastructure apps for software-defined networking, software-defined storage, cybersecurity, telemetry and in-network computing applications yet to be invented.”

Huang is also on a partnership with VMware, which it announced last week for porting VMware onto BlueField.

Huan opined, “VMware "runs the world's enterprises — they are the OS platform in 70 per cent of the world's companies. BlueField-2 is sampling now, BlueField-3 is finishing and BlueField-4 is in high gear, Huang informed. We are going to bring a ton of technology to networking. In just a couple of years, we'll span nearly 1,000 times in compute throughput on the DPU".

Huang announced, “Microsoft is adopting NVIDIA AI on Azure to power smart experiences on Microsoft Office, including smart grammar correction and text prediction.”

Using Nvidia GPUs for inference, Microsoft Office joins Square, Twitter, eBay, GE Healthcare and Zoox.

Huang said, “Cloudera, a hybrid-cloud data platform that lets customers manage, secure, analyze and learn predictive models from data, will accelerate the Cloudera Data Platform with Nvidia RAPIDS, Nvidia AI and NVIDIA-accelerated Spark.”

Nvidia and VMware further announced a second partnership to create a data center platform for supporting the GPU acceleration of all the three major computing domains – virtualized, distributed scale-out and composable microservices.

Huang stated that an early access program for a new service called Nvidia Fleet Command.

 The new application by Nvidia Fleet Command will make it easy for deploying and managing updating the IoT devices, combining the security and real-time processing capabilities of edge computing with the remote management and ease of software-as-a-service.