Feature Description

The current in-service upgrade strategy only supports upgrading one node at a time. For bigger clusters, more than six nodes, this upgrade strategy leads to longer upgrade periods, which mostly exceed the maintenance window (MW) limits.

This feature enables you to perform parallel upgrades for multiple nodes concurrently for faster in-service upgrades without impacting the availability and replication for any NF.

Architecture

This following images show the high-level design of the group upgrade flow for K8s and KVMs.
Upgrade Flow for K8s Clusters
Upgrade Flow for KVM Clusters