Omnia: Overview

Omnia, deriving its name from the Latin term denoting “all” or “everything”, is a deployment tool tailored to configure Dell PowerEdge servers operating on standard Linux OS images into clusters capable of handling HPC, AI, and data analytics workloads. Leveraging Slurm, Kubernetes, and complementary packages, it orchestrates job management and enables execution of varied workloads on the same converged solution. Omnia is a collection of open-source Ansible playbooks, continually evolving to accommodate a wide array of workloads effectively.

• Architecture
• Omnia Stack
• New Features
  • BuildStreaM Pipeline Architecture and API Enhancements
  • Vector Telemetry Pipeline for Data Routing
  • PowerScale Telemetry for Storage Monitoring
  • Vast Repo and Vast Client Installation
  • Minimal OS Functional Groups
  • NVIDIA DCGM and CUDA Toolkit Provisioning for Slurm GPU Nodes
  • NVIDIA HPC SDK Provisioning for Slurm Clusters
  • One-Shot Combined Log Extraction for Debugging
  • BMC Discovery via Dell OpenManage Enterprise
• Release Information
  • Version 2.1.0.0
  • Version 2.0.0.0
• Types of Clusters Deployed by Omnia
  • Service Kubernetes Cluster
  • Slurm Cluster
• Network Topologies
  • Network Topology: Dedicated Setup
  • Network Topology: Shared LOM Setup
  • Network Topology: Hybrid Setup
• Blogs about Omnia

If you have any feedback about Omnia documentation, please reach out at omnia.readme@dell.com.