Nvidia announced the release of several AI computing initiatives for enterprise AI product development and operation on its online-only Computex 2021. Nvidia used the opportunity to promote a cloud-hosted Base Command AI software development platform and several AI-optimized x86 servers as part of its Nvidia-certified systems program.
The conference was canceled in 2020 due to the outbreak of the COVID-19 pandemic. This year, it’s back to full steam, hosting the biggest names in the computing industry, including Nvidia, AMD, Intel, Acer, Arm, and Gigabyte.
Nvidia explained the new announcements would allow different enterprises to power their AI workloads on traditional data center infrastructure.
They partnered up with NetApp to develop the new cloud-hosted Base Command AI software development platform. So, it provides users with Nvidia DGX SuperPOD AI supercomputers and NetApp data management tools.
To be more specific, Nvidia developed the BCP featuring integrated monitoring and reporting dashboards, with command-line APIs and AI developer tools, such as Nvidia’s NGC catalog of AI, analytics software, APIs for integrating with MLOps software, and Jupyter notebooks.
The Base Command Platform (BCP) accelerates the delivery of AI projects. The Cloud AI allows researchers and data scientists to work on the same project simultaneously, maximizing the enterprise’s productivity. In other words, it provides a way to move AI projects from prototypes to production quickly.
NVIDIA-Certified servers ensure that modern applications can run at maximum performance almost anywhere—from #datacenter to edge. Watch to learn how.
— NVIDIA Data Center (@NVIDIADC) May 5, 2021
That’s why it’s appropriate for large-scale, multi-user, and multi-team AI development workflows. Nvidia made the platform available at a $90,000 monthly subscription for early customers in their press briefing.
As part of the Nvidia-certified systems program, the new AI-optimized servers come from major computer manufacturers, including Dell Technologies, Gigabyte, Asus, or Lenovo.
Manuvir Das, who serves the role as head of Enterprise Computing at Nvidia, had this to say:
The open, growing ecosystem of Nvidia-Certified Systems provides unprecedented customer choice in servers validated by Nvidia to power world-class AI.
He also added:
The new servers will become some of the highest-volume x86 servers used in mainstream data centers, bringing the power of AI to a wide range of industries, including health care, manufacturing, retail, and financial services.
While the new x86 servers based on Nvidia Ampere architecture GPUs are available now, they will release the ones using BlueField-2 DPUs later this year. When it comes to the non-x86 machines powered by Arm CPUs, we’ll have to wait until 2022.
Nvidia added that they expect the new systems to work with its AI Enterprise software, VMware vSphere, Nvidia Omniverse Enterprise, Red Hat OpenShift, and Cloudera.