Would you like some AI with your hybrid cloud? Perhaps with a side helping of edge computing?
As enterprise customers wrestle with new computing architectures, NVIDIA wants to make it easier to access its powerful artificial intelligence hardware across their infrastructure. Today NVIDIA is rolling out services that integrate its AI technology into hybrid cloud deployments, including a new software service for managing on-demand AI workloads across a distributed edge network.
The first data center partner for the NVIDIA AI Launchpad service will be Equinix, the largest player in colocation and interconnection. Starting later this year, customers will be able to rent AI servers at Equinix on several types of NVIDIA-certified hardware, including AI infrastructure as a service (IaaS) hybrid clouds using bare metal servers. The AI Launchpad service will be available as an on-demand service with hourly billing, and the option to reserve servers for longer term usage.
“NVIDIA AI LaunchPad puts AI at the fingertips of enterprises everywhere with fully automated, hybrid-cloud infrastructure and software for every stage of the AI lifecycle,” said Manuvir Das, head of Enterprise Computing at NVIDIA.
Enterprise users are eager to use AI to enhance their products, services and IT operations. But AI requires powerful hardware and large datasets, and organizations must sort out the best way to deploy these capabilities across infrastructures that may include on-premises data centers, public cloud platforms and cloud workloads.
NVIDIA has worked closely with the data center industry to support its high-density compute hardware. With 220 data centers, Equinix offers a large global footprint for AI Launchpad.
Hybrid Cloud as an AI Platform
NVIDIA executives said the AI infrastructure at Equinix will enable customer to deploy in minutes to provide enterprises with access to a wide spectrum of NVIDIA resources that support virtually every aspect of AI, from data center training and inference to full-scale deployment at the edge.
That will include NVIDIA Fleet Command, a managed platform to deploy edge AI services across a distributed network. Fleet Command is remote management software that lets users install and manage software from a central location.
“The ability to deploy and manage AI applications at the edge is one of the most complex problems facing businesses as they move toward an Internet of Things era, including smart factories, intelligent retail and smart cities,” said Das. “NVIDIA Fleet Command drives efficiency across a business, helping scale edge AI with unprecedented speed.”
Equinix is the world’s leading provider of colocation, with a focus on interconnections that move data traffic across network boundaries. This formula has allowed the company to create a digital infrastructure empire, with a presence in more than 50 markets across the globe.
“Many industries use private clouds to keep costs down by having computing resources close to the data, for performance, data privacy, ownership and sovereignty reasons,” said Steve Steienhilber, vice president, Business Development at Equinix. “With NVIDIA AI LaunchPad globally available at Equinix, enterprises will have immediate access to NVIDIA software and NVIDIA-Certified infrastructure in a comprehensive hybrid-cloud solution.”
NVIDIA: We Want to Democratize AI
NVIDIA’s GPU technology has been one of the biggest beneficiaries of the rise of specialized computing, gaining traction with workloads in supercomputing, artificial intelligence (AI) and connected cars. NVIDIA has been investing heavily in innovation in AI, which it sees as a pervasive technology trend that will bring its GPU technology into every area of the economy and society.
This week’s service offerings are part of NVIDIA’s push to make its technology available to a broader set of customers, including those who may not have massive budgets or data center infrastructures that can support high-density deployments.
“AI is the superpower, that’s going to be part of everybody’s life,” said Justin Boitano, the VP/GM for Enterprise and Edge Computing at NVIDIA. “It’s time to democratize AI across every industry, by bringing its transformational power to every company, and every company’s customers.
“The most common question we get in a supply-constrained world is ‘how do I get access to infrastructure to get started right away,’ ” said Boitano. “We want to provide that infrastructure with providers like Equinix. This program makes it easier for enterprises to get access to AI infrastructure in a cloud-like consumption model.”
NVIDIA works with about 20 data center operators through the DGX Ready program, which certifies providers that can support deployments of its DGX ‘supercomputer in a box.” Not every data center can support these densities, especially on-premises facilities which were likely designed and built in an era of much lower rack power densities.
DGX SuperPods, then Edge Servers to Follow
The first phase of the NVIDIA AI Launchpad Partnership with Equinix will focus on the DGX platform. Equinix will be the first colocation provider with an NVIDIA DGX SuperPOD offered for rental with Base Command, the NVIDIA platform for managing AI development. The SuperPOD combines multiple DGX A100 systems, which take up 6 rack units.
Access to the DGX SuperPOD is expected to be available by the end of summer, and NVIDIA has previously said that pricing for the Base Command Platform with DGX SuperPOD service starts at $90,000/month. The companies have not outlined the full specs for the SuperPOD deployment at Equinix.
In the next phase, customers will have on-demand access to AI capabilities on bare metal deployments of NVIDIA-certified servers from Dell Technologies and Lenovo. These servers are part of the NVIDIA EGX Edge AI platform, which is designed to offer distributed computing horsepower for hospitals, stores, farms and factories that must manage growing amounts of data streaming from edge sensors. The platform makes it possible to securely deploy, manage and update fleets of servers remotely.
Users will be able to add NVIDIA Base Command or Fleet Command services to these servers, providing options for either development or production workloads. “They can on-board our services, and we’re working to make it as seamless and unified as possible,” said NVIDIA’s Boitano. “Even if it’s delivered at Equinix, it feels like one common consumption model.”
The most powerful opportunity will be to deploy AI servers across edge locations. NVIDIA says it is working with Equinix on the timing and locations where AI Launchpad will be available as an edge service. Pricing for additional NVIDIA services will be available as each reaches general availability at Equinix.