VMware and NVIDIA announced a broad partnership to offer an end-to-end enterprise platform for AI and a new architecture for data center, cloud and edge that uses NVIDIA® DPUs (data processing units) with the latest support and applications generation.
With the collaboration, the AI software suite available on the NVIDIA NGCTM hub will be integrated with VMware vSphere, VMware Cloud Foundation and VMware Tanzu. This will help accelerate the adoption of Artificial Intelligence, allowing companies to extend their existing infrastructure to AI, manage all applications with a single set of operations and implement an environment where the data resides, in the data center, cloud and edge.
In addition, as part of the Monterey Project announced separately, companies will also partner to deliver a hybrid cloud architecture based on SmartNIC technology, including NVIDIA's programmable BlueField®-2. The combination with the VMware Cloud Foundation will offer next-generation infrastructure developed specifically for the demands of AI, machine learning, high throughput and data-centric applications. The agreement will also provide expanded application acceleration beyond AI for all corporate workloads and promote an extra layer of security through a new architecture that offloads critical data center services from the CPU to programmable SmartNICs and DPUs.
“We signed a partnership with NVIDIA to bring AI to all companies; a true democratization of one of the most powerful technologies, ”says Pat Gelsinger, CEO of VMware. “We are also collaborating to define a new architecture for the hybrid cloud - a purpose developed to support the needs and demands of the next generation of applications. Together, we are positioned to help all companies accelerate the use of innovative applications to boost their business. "
"AI and machine learning have quickly evolved from research labs to data centers in companies from virtually every industry and geography," says Jensen Huang, founder and CEO of NVIDIA. “NVIDIA and VMware will help customers turn each corporate data center into an accelerated AI supercomputer. NVIDIA's DPUs will give companies the ability to build secure, programmable, software-defined data centers that can accelerate all enterprise applications with exceptional value, ”he adds.
AI-ready enterprise platform
The first aspect of collaboration between NVIDIA and VMware - the integration of NVIDIA NGC with VMware vSphere and VMware Cloud Foundation - will simplify the deployment and management of AI for the most demanding workloads. Industries ranging from healthcare to financial services, retail and manufacturing will be able to easily develop and deploy AI workloads using containers and virtual machines, on the same platform as their enterprise applications, at scale in the hybrid cloud.
VMware customers will also be able to accelerate data science and AI workloads based on their existing infrastructure, resources and tool sets - helping to expand the adoption of AI and ML (machine learning) technologies. Data scientists, developers and researchers will have immediate access to the wide range of cloud-native and GPU-optimized NGC-specific containers, models and software development kits. The NGC software is compatible with a select set of pre-tested NVIDIA A100 servers and from leading system manufacturers such as Dell Technologies, Hewlett Packard Enterprise (HPE) and Lenovo.
Offering new hybrid cloud architecture for next generation applications
The second element of the collaboration between VMware and NVIDIA recognizes that, as the next generation workloads increase in complexity, SmartNICs and DPUs are critical technologies to securely accelerate a wide range of enterprise applications where the data resides.
The partnership will deliver a new hybrid cloud architecture that will help organizations evolve their infrastructure and operations, in addition to introducing a new security model that frees up hypervisor, network, security and CPU storage tasks for the DPU. This new architecture will also extend the VMware Cloud Foundation operating model to bare metal servers.
EN 
PT 
