Nvidia and VMware have expanded an existing partnership to support Nvidia’s AI Enterprise software suite in VMware vSphere cloud environments that use VMware’s Tanzu Kubernetes container clustering platform. This will result in faster deployments of AI tools across existing clouds to where enterprises need them the most, the companies said.
During VMware’s VMWorld 2021 event this week, the companies announced that enterprises using Nvidia’s software can now obtain a license to run trials of their AI projects using their existing vSphere with Tanzu infrastructures. This represents the latest step in a partnership announced by the two companies just over a year ago, and which initially focused on supporting Nvidia AI Enterprise in vSphere before progressing to Tanzu and other VMware platforms.
“Over the course of the past year we have been able to deliver more broadly on the promise of AI-ready enter[rise platforms in partnership with VMware,” said John Fanelli, vice president of enterprise product management at Nvidia, adding that support for Tanzu Kubernetes container clusters directly addresses how large enterprises are working with AI applications.
“This is one way that we’re helping bring AI more quickly into the enterprise environment, where we see a need for resiliency and quality of service,” added Lee Caswell, vice president of product marketing at VMware.
Project Monterey
Also during this week’s event, Nvidia said it is expanding its early access program for VMware’s Project Monterey, which is being hosted by Nvidia and is focused on putting BlueField data processing units in enterprise data centers. to include Lenovo. Project Monterey is an effort to bring higher performance and security to those data centers by off-loading workloads from CPUs to DPUs.
“In data centers, CPUs have been running application workloads that take up valuable CPU resources, and we now want to offload those to a new instance called the DPU,” Caswell said. “A DPU acts independently as an air-gapped method of applying infrastructure management and releasing CPU resources to run what they were designed for.”
Security applications are one of the most likely candidates to get moved to DPUs because enterprises recently have upped their investments in security solutions, but may not have clear strategies for managing these applications “at an intrinsic level,” Caswell added. “Eventually, you’ll see timelines from us as we prepare for Project Monterey moving to product.”
RELATED: Nvidia partners with VMware for ease in managing AI infrastructure