NVIDIA Cloud Accelerator Documentation

NVIDIA Cloud Partner Software Reference Guide

Contents

NCP Software Reference Guide

• Introduction
  • Document Structure
  • Audience and Scope
  • User Personas
  • Terms and Definitions
• Data Center Architecture
  • GPU Compute Node
  • Networking
  • GPU Compute Rack
  • Storage
  • Data Center View

Part 1: Software Reference Guide

• NCP Software Reference Guide
  • Tenant Compute View
  • Operator View
• Infrastructure-as-a-Service
  • IaaS Architecture
    • Component Descriptions
    • Network Management
    • Compute Management
    • Storage Management
  • NVIDIA Software for Infrastructure-as-a-Service
• Container-as-a-Service
  • Kubernetes
    • Capabilities Required
    • K8s-Native ML/AI Frameworks and Tools
    • Kubernetes Architecture for ML/AI
  • NVIDIA Software for Container-as-a-Service
    • Container Platform
    • Kubernetes Operators
• AI Platform-as-a-Service
• Break-Fix Architecture
  • Break-Fix Overview
  • Break-Fix
    • Node-Level Health Checks
• Performance Requirements
  • Virtual Machine Networking
  • Virtual Machine Traffic Acceleration
  • Virtualizing a GPU
  • Enabling Multitenancy and Isolation
  • Storage Connectivity
  • Virtual Storage
• Telemetry and Observability
• Workload Isolation
  • Boot and Attestation
    • Secure Boot
    • Measured Boot
    • Remote Attestation
• Shared Responsibility Model

Part 2: Software Components

• NVIDIA Software Components
  • Key Software Components
• NVIDIA Software for Infrastructure as a Service
  • Network Management
    • NVIDIA Unified Fabric Manager (UFM)
    • NVIDIA User Experience (NVUE)
    • NVIDIA NetQ
    • NVIDIA NMX
    • DOCA / HBN
  • Compute Management
    • NVIDIA Bare Metal Manager
    • Virtualization
    • NVIDIA vGPU for Compute
    • Base Command Manager
    • Base Command Manager with Virtual Machines
    • NVIDIA Mission Control
  • Software References for Storage Management
    • GPUDirect Storage (GDS)
    • GPUDirect RDMA (GDR)
• NVIDIA Software for Container as a Service
  • Provisioning Kubernetes
  • Dynamic Resource Allocation (DRA)
  • IMEX (Multi-Node NVLink)
  • GPU Operator
  • Container Toolkit
  • Network Operator
• NVIDIA Software References for AI Platform
  • Run:ai
  • NVCF
  • Slurm
  • NeMo