Dell’s Omnia Automates HPC, PowerEdge Gets Nvidia Refresh
Dell Technologies today announced the general availability of Omnia, its open source platform for managing high-performance computing (HPC) and artificial intelligence (AI) workloads.
Omnia is an open source set of Ansible Playbooks developed by Dell’s HPC and AI innovation team in partnership with Intel and the broader HPC community. If this sounds familiar, Dell quietly released Omnia on Github earlier this year.
The platform automates the process of downloading and installing open source software for things like cluster management, application execution, accelerator drivers, frameworks, applications and platforms, and it manages systems running Slurm or Kubernetes, explained Caitlin Gordon, VP of product management at Dell Technologies.
“It’s all about simplifying and easing the development and the management of high-performance clusters for not just HPC but AI,” she said.
And because these automations are built on Ansible Playbooks, “you can compose and recompose the stack, based on what you need,” Gordon said. “We’re really taking an infrastructure as code approach here.”
This, she added, allows users to compose HPC deployments around a wide variety of workloads and hardware from multiple vendors. “Of course, because its open source, it can be shaped to whatever your needs are,” Gordon said. “We’ll continue to integrate the latest open source tools, but we also, even more importantly, have the collective power of the HPC community already working behind this.”
Arizona State University, one of Dell Technology’s HPC and AI centers for excellence, is among the first to deploy Omnia and begin contributing to its development, according to Gordon.
Dell HPC Nabs VMware Support, PowerEdge Gets Nvidia Upgrade
Dell Technologies also extended support for VMware to its HPC on-Demand program, which launched last year.
“Being able to consume things as a service be able to access resources as you need them, be able to move faster and simply pay for what you’re using. That’s really what HPC on demand is all about,” Gordon said. “It’s really about being able to support our customers who may not have the resources, or may not want to invest in their own infrastructure and people, because their HPC workloads maybe are inconsistent, and/or maybe they need to complement what they already have in their environment with a secondary environment to burst to.”
With support for VMware Cloud Foundation, VMware Cloud Director, and VMware vRealize Operations, customers can now extend these capabilities beyond single-tenant environments to include multi-tenant use cases.
Meanwhile, Dell Tech is adding Nvidia A10 and A30 GPUs to its PowerEdge server portfolio. “These are really ideal for things like VDI inferencing and interactive rendering,” Gordon said.