Dell HPC and AI work generates Omnia software for easy deployment
Dell Technologies this week showcased the open source Omnia software to highlight its latest initiatives to make it easier to deploy and manage AI, analytics, and high performance computing applications.
The Dell Technologies HPC & AI Innovation Lab collaborated with Intel and the HPC community on Omnia software to help users automate the provisioning and management of AI and compute-intensive workloads by creating a single pool of flexible resources to meet changing needs.
Caitlin Gordon, vice president of product management for integrated solutions at Dell Technologies, said Omnia would allow users to use the same infrastructure for different types of tasks and put the “right software” on each server. depending on the use case, such as HPC simulations. or neural networks for AI.
“You will be able to reduce your deployment time from a few weeks to a few hours,” Gordon said. “So it’s all about flexibility. One of the coolest things about Omnia is that you can compose and recompose the stack to suit your needs, really taking an infrastructure-as-code approach here. “
The Omnia software stack combines a set of Ansible playbooks for faster deployment, the Kubernetes container orchestration engine, Slurm workload manager, library frameworks, services, and applications. Gordon said the Omnia software is available for a few months on GitHub. Contributors to the open source Omnia project include Arizona State University Research Computing.
Focus on HPC, AI and Dell Analytics
Addison Snell, CEO of Intersect360 Research, said Dell’s version of Omnia matches market trends in diversification, including workloads blending AI, data analytics, and traditional science and technology elements. of HPCs such as specialized processors and accelerators.
“Omnia is designed to be a common and consolidated environment to facilitate the deployment and management of clusters through these growing needs. By making it open source, they allow for customization, like a sandwich shop that allows a customer to change their selection, ”Snell said. “In this way, Omnia is also following the trend that we see towards composability, allowing the selection of different modules according to the tasks at hand, including through the use of containers. “
Snell said he expects Omnia to appeal the most to organizations looking for supported environments that can cover a wide range of high-performance workloads. He said the Intersect360 polls show that 80% of HPC Users have either integrated machine learning into their environments or are actively working to do so, usually on a shared or common infrastructure. Dell designed Omnia to target these types of environments, Snell said.
Steve McDowell, senior technology analyst at Moor Insights & Strategy, said it was not surprising to see Dell release Omnia software to facilitate deployments to its servers and storage, as the company has grown increasingly focused on HPC and AI for the past few years.
“HPC, supercomputing, and AI are all workloads that treat infrastructure as fungible, which means the underlying hardware is reassigned and reconfigured to meet the needs of the application,” McDowell said. . “The industry has widely adopted Ansible as a mechanism to dynamically configure and reconfigure infrastructure. It is natural that Dell offers these capabilities as Ansible playbooks.
McDowell said Dell is not alone in taking the Ansible approach. He noted that IBM and HPE, for example, support the HPC and AI community with Ansible playbooks.
Dell HPC on demand expands to VMware
Also as part of this week’s ISC High Performance 2021 Digital event, Dell has extended its on-demand HPC offering to VMware’s cloud foundation, Cloud Director, and vRealize Operations. HPC on demand is designed to give customers quick access to resources and allow them to pay only for what they use.
Gordon said adding VMware support to Dell HPC On-Demand would give Dell the ability not only to offer HPC on-demand in a single-tenant environment, but also in multi-tenant environments.
“It’s really about being able to support our customers who may not have the resources or want to invest in their infrastructure and people because their HPC workloads may be inconsistent. Or maybe they need to supplement what they already have in their environment with a secondary environment to burst into, ”Gordon said.
McDowell said traditional HPC management tools are often not the same ones businesses use to manage their IT infrastructure. Thus, adding VMware as a management option for Dell’s on-demand supercomputing offering should make it easier for traditional enterprise IT administrators who are already comfortable with VMware to easily manage workloads. of HPC work, he said.
Also this week, Dell announced that it now offers options for Nvidia A30 and A10 Tensor Core. GPU with Dell EMC PowerEdge R750, R750xa, and R7525 servers to handle AI and deep learning workloads. Gordon said ideal use cases include VDI, inference, and interactive rendering.
Carol Sliwa is a senior writer at TechTarget and covers storage arrays and drives, flash and memory technologies, and enterprise architecture.