Edge Site Management Strategy¶

Overview¶

Large-scale deployments often span multiple geographic regions, data centers, and edge locations. HPE Morpheus Software supports a three-tier management architecture that scales from a single appliance to a fully federated multi-region deployment:

Architecture¶

              ┌─────────────────────────┐
              │    Morpheus Central      │
              │  (Federated Management)  │
              └────────┬────────┬────────┘
                       │        │
          ┌────────────┘        └────────────┐
          │                                   │
┌─────────▼──────────┐             ┌─────────▼──────────┐
│  Regional Manager  │             │  Regional Manager  │
│   (Region A)       │             │   (Region B)       │
└──┬──────┬──────┬───┘             └──┬──────┬──────┬───┘
   │      │      │                    │      │      │
   ▼      ▼      ▼                    ▼      ▼      ▼
┌─────┐┌─────┐┌─────┐             ┌─────┐┌─────┐┌─────┐
│Edge ││Edge ││Edge │             │Edge ││Edge ││Edge │
│Wrkr ││Wrkr ││Wrkr │             │Wrkr ││Wrkr ││Wrkr │
└─────┘└─────┘└─────┘             └─────┘└─────┘└─────┘

Tier 3: Distributed Workers at the Edge¶

Distributed Workers are the foundation of edge site management. Each Worker is a lightweight service deployed at a remote or edge location that maintains an outbound connection to its regional HPE Morpheus Software Manager. Workers proxy:

• Cloud API traffic — Communicates with local hypervisors (VMware, HVM, Nutanix, SCVMM, Hyper-V, OpenStack, etc.) on behalf of the regional Manager

• Agent relay — Relays HPE Morpheus Software Agent traffic from managed VMs back to the Manager

• Console gateway — Routes browser-based console sessions to local VMs without requiring direct connectivity from user browsers to the hypervisor network

• HVM quorum witness — Provides witness services for two-node or stretch HVM clusters at edge sites

When to deploy a Worker at an edge site:

• The regional Manager cannot directly reach the hypervisor management network at the edge location

• Agent traffic from edge VMs cannot route directly to the Manager

• Users need console access to VMs at edge sites without VPN

• Network segmentation or firewall policy requires all traffic to egress through a known endpoint

• HVM clusters at edge sites need a remote quorum witness for availability

Deployment considerations:

• Workers require only outbound HTTPS (TCP 443) to the regional Manager

• Multiple Workers can share the same Worker key for HA at a single edge site

• A single Worker can serve multiple roles (cloud proxy, console gateway, witness)

• Separate Workers per role may be preferred for fault isolation or independent scaling

For installation, configuration, and HA deployment details, see Distributed Workers. For two-Host GFS2 witness operations, see Two-Node Clusters with a Witness. For stretch-cluster witness operations, see Stretch Clusters & Witness Nodes.

Tier 2: Regional Morpheus Managers¶

Each region, data center, or major campus operates its own HPE Morpheus Software Manager appliance (or HA appliance cluster). The regional Manager:

• Owns all cloud integrations, groups, and tenants for its scope

• Handles provisioning, automation, governance, and policy enforcement

• Manages its own set of Distributed Workers deployed at edge locations within the region

• Operates independently if connectivity to Morpheus Central is interrupted

Sizing guidance:

• A single Manager can manage thousands of VMs across many Distributed Workers

• Deploy additional regional Managers when geographic latency, compliance boundaries, or blast-radius requirements demand independent control planes

• Each Manager maintains its own database, elasticsearch, and RabbitMQ (or HA equivalents)

For appliance installation and HA configuration, see /getting_started/ha/ha.

Tier 1: Morpheus Central (Federated Management)¶

Morpheus Central provides a federated management layer that connects multiple regional Managers into a single organizational view:

• Unified visibility — View infrastructure, instances, and operations across all registered regions from a single interface

• License management — Centralized license allocation and consumption tracking across all Managers

• Cross-region coordination — Federated reporting, compliance visibility, and operational awareness

Each regional Manager registers with Morpheus Central via Administration > Settings (Morpheus Central tab). Registration requires outbound HTTPS connectivity from the Manager to the Central service.

Deployment Patterns¶

Single Region with Edge Sites¶

The simplest pattern: one Manager with multiple Workers at remote sites.

• 1 HPE Morpheus Software Manager (data center or central office)

• N Distributed Workers (one per edge site or remote network zone)

• Workers proxy cloud and agent traffic from edge hypervisors

Multi-Region with Edge Sites¶

For organizations spanning multiple geographies:

• 1 Morpheus Central instance

• N HPE Morpheus Software Managers (one per region)

• M Distributed Workers per region (at edge locations within each region)

This pattern provides:

• Regional autonomy (each Manager operates independently)

• Edge reach (Workers extend each Manager to remote sites)

• Federated oversight (Central provides cross-region visibility)

Compliance-Driven Segmentation¶

For environments with strict data sovereignty or compliance boundaries:

• Separate Managers per compliance zone (e.g., one for EMEA, one for US)

• Workers within each zone cannot communicate with Managers in other zones

• Central provides visibility across zones without moving workload data across boundaries

Planning Checklist¶