Application Instances on Nodes and Clusters

​

Introduction

Applications in Edge Infrastructure Services are deployed instances of edge app images running on nodes or clusters. Each application instance represents a single workload (a virtual machine, container, Docker Compose stack, or Azure IoT Edge module) deployed to a specific node. The Applications area is where you create new application instances, monitor their status, and manage their lifecycle. Applications were previously called Edge App Instances.

​

Prerequisites

• You have an active Edge Infrastructure Services enterprise.
• You have the SysManager or SysAdmin role in your enterprise. See Manage Roles.
• At least one edge app image exists in the Build section.
• At least one node is onboarded and online.

​

How It Works

An application instance is created by deploying an edge app image from the Build section to a target node or cluster. The image defines the workload type and configuration; the instance represents that workload running on a specific node. Each instance has a status that reflects its current state on the node. The status is reported by EVE-OS (Edge Virtualization Engine) running on the node and updated automatically in the platform.

​

Application instance statuses

For a description of every run state, see Application Instance Status and Health.

​

Deployment targets

Application instances can be deployed to:

• A single node: the workload runs on one specific node.
• A cluster: the workload is scheduled across the cluster. You can optionally pin the instance to a preferred node and configure failover behavior.

​

Key Benefits

• Centralized lifecycle management. Activate, deactivate, restart, update, and delete application instances from a single location without connecting directly to the node.
• Remote access. Open a VNC remote console or SSH terminal to any online application instance through EdgeView without requiring direct network access to the node.
• Snapshot and rollback. Create snapshots of running instances and roll back to a previous state if an update causes problems.
• Patch envelope delivery. Attach runtime bundles to push file-level updates to running instances without redeploying the application.
• Real-time monitoring. View logs and events for each application instance directly from the instance detail page.

​

Next Steps