Is this page helpful?

Upgrading from 7.7 to 7.8#

Use this checklist when upgrading existing 7.7 deployments to 7.8. Run each step in a maintenance window and validate before proceeding.

Component Version Delta#

Compare pinned infrastructure versions that change between 7.7 and 7.8 before you upgrade. Components that keep the same version are omitted. For full component descriptions, refer to the 7.8 Release Notes. For platform and OS support, refer to the 7.8 Support Matrix.

Table 120 7.7 to 7.8 Component Version Delta#

Component

7.7

7.8

NVIDIA Data Center GPU Driver

580.173.02

580.178.04

NVIDIA Fabric Manager

580.173.02

580.178.04

NVIDIA Virtual GPU Manager

580.159.01

580.178.05

NVIDIA vGPU for Compute Guest Driver (Linux)

580.173.02

580.178.04

NVIDIA vGPU for Compute Guest Driver (Windows)

582.70

582.78

NVIDIA Run:ai

2.25

2.26

NVIDIA Network Operator

26.4.0

26.4.1

NVIDIA NIM Operator

3.1.1

3.1.2

Upgrade Sequence#

Complete the following steps in order during a maintenance window. Validate each step before you continue to the next.

Table 121 7.7 to 7.8 Upgrade Sequence#

Step

Action

Reference

  1. Review

Read the component version delta in the table above and review per-component release notes for breaking changes, deprecations, and feature additions.

7.8 Release Notes

  1. Validate

Confirm hardware, hypervisor (for virtualized deployments), and operating-system compatibility against the 7.8 support matrix.

7.8 Support Matrix

  1. Plan rollback

In virtualized deployments, snapshot or back up VMs and capture current driver, operator, and licensing configuration before upgrading. Document the rollback path for each component.

N/A

  1. Upgrade host components

Depending on your deployment type (bare metal or virtualized), upgrade the NVIDIA Data Center GPU Driver (applies to bare metal deployments), Virtual GPU Manager (only applies to vGPU for Compute deployments), and DOCA Driver. Fabric Manager is included in the NVIDIA AI Enterprise drivers and updates with the GPU Driver package.

Installing NVIDIA vGPU for Compute

  1. Upgrade Kubernetes operators

Upgrade NVIDIA GPU Operator, Network Operator, NIM Operator, DPU Operator (DPF), and NVIDIA Run:ai (self-hosted) to the versions in the component version delta table above. NVIDIA Run:ai SaaS upgrades are managed by NVIDIA and require no customer-side upgrade.

Component version delta above

  1. Upgrade guest VMs

For virtualized deployments, upgrade NVIDIA vGPU for Compute Guest Driver and Container Toolkit in tenant VMs.

Installing the vGPU Guest Driver

  1. Verify licensing

Confirm each licensed vGPU VM reaches the NVIDIA License System and shows Licensed status.

Verifying License Configuration

  1. Validate workloads

Run a representative CUDA or AI/ML workload to confirm performance, feature parity, and operator-managed scheduling behavior.

N/A

For lifecycle policy, branch support windows, and migration windows beyond 7.x, refer to the NVIDIA AI Enterprise Lifecycle Policy.