Dell Technologies has announced the release of Omnia, a new open-source software platform for managing the intersection of high-performance computing, artificial intelligence, and data analytics.
Additionally, it is expanding its support for EMC PowerEdge server accelerators to assist organisations with data-intensive workloads such as genome sequencing and product development simulations.
Omnia was developed at Dell Technologies’ High Performance Computing and Artificial Intelligence Innovation Lab in collaboration with Intel and with support from the HPC community. The open-source software is intended to automate the provisioning and management of HPC, AI, and data analytics workloads, thereby establishing a centralised pool of resources capable of meeting growing and diverse demands.
The Omnia software stack is an open-source collection of Ansible playbooks for deploying converged workloads with Kubernetes and Slurm, as well as library frameworks, services, and applications. Omnia automatically imprints a software solution on each server based on the use case, such as high-performance computing simulations, artificial intelligence, or memory graphics processing for data analytics, reducing deployment time from weeks to minutes.
“We are committed to enabling innovation,” says Chris Kelly, vice president of data centre and compute at Dell.
“Our high-performance computing and artificial intelligence solutions enable renowned organisations in the Asia Pacific and Japan regions, such as Kyoto University and AI Singapore, to handle data-intensive tasks such as advanced research. Globally, as AI, HPC, and data analytics continue to converge at a breakneck pace, IT teams are confronted with siloed storage and networking configurations, as well as the challenge of providing the necessary technology resources to meet shifting demands.
“With the release of Dell’s open-source Omnia software, teams can significantly simplify the management of advanced computing workloads, enabling them to accelerate research and innovation,” he says.
Dell believes that community involvement and contribution are critical to Omnia’s growth. Additionally, the Dell Technologies HPC & AI Innovation Lab collaborated closely with Arizona State University Research Computing on the development of Omnia in order to improve support for mixed workloads such as simulation, high-throughput computing, and machine learning.
“ASU and Dell Technologies engineers collaborated on the development of Omnia,” says Douglas Jennewein, senior director of research computing at Arizona State University.
“Working on code that will simplify the deployment and management of these complex mixed workloads at ASU and throughout the advanced computing industry has been a rewarding endeavour.”
Dell EMC PowerEdge R750, R750xa, and R7525 servers now support NVIDIA A30 and A10 Tensor Core GPUs. NVIDIA A30 GPUs are optimised for a wide variety of AI inference and enterprise compute workloads, including conversational AI and computer vision. Additionally, customers can run mixed AI and graphics workloads on a single infrastructure with NVIDIA A10 GPUs, which is ideal for deep learning inference and computer-aided design.
