# Scaling to the Millions: The Logic of Fleet Sharding

Architectural Purpose and Scaling Intent 
-----------------------------------------

This document describes the architectural foundation used to scale Unified Endpoint Management for **very large enterprise fleets**, including environments managing **hundreds of thousands of devices**.

The architecture is built on two complementary principles:

- **Decentralization for performance and resilience**
- **Centralized orchestration for visibility and governance**

Together, these principles allow global organizations to operate regionally distributed device fleets as a single logical system without introducing performance bottlenecks or regulatory risk.

Regional Node Segmentation and Fleet Sharding 
----------------------------------------------

### Why Fleet Sharding Exists 

As device counts grow, monolithic backends become fragile. Database contention increases, user interfaces slow down, and failures propagate widely. To avoid this, large fleets are divided into **independent operational shards**.

### How Sharding Is Implemented 

Enterprise fleets exceeding **500,000 devices** are segmented into **10 to 20 autonomous nodes**, each functioning as a self contained tenant.

Each node manages its own:

- Device records
- Policy execution
- Compliance evaluation
- Telemetry ingestion

There is no shared runtime dependency between nodes.

### Operational and Technical Outcomes 

**Performance Isolation** Each shard maintains its own database backend. This significantly reduces IOPS pressure and ensures [device actions](https://www.hexnode.com/mobile-device-management/help/remote-actions-supported-by-hexnode-a-platform-wise-comparison/) and [compliance checks](https://www.hexnode.com/mobile-device-management/help/compliance-enforcement-and-remediation/) consistently complete within millisecond level response times.

**Regional Responsiveness** Nodes are deployed geographically close to administrators. This minimizes Time to First Byte and keeps the management console responsive regardless of overall fleet size.

**Fault Containment** Maintenance events or service degradation affecting one node do not impact others. Operational integrity is preserved across the remaining shards.

Global Data Sovereignty and Regional Compliance Alignment 
----------------------------------------------------------

### Data Residency as a First Class Design Constraint 

The architecture is explicitly designed to meet international data residency and privacy regulations by **pinning data to geographic regions** rather than abstracting it away.

Device metadata and personally identifiable information remain within legally required boundaries at all times.

### Region Specific Deployment Strategy 

**European Union** Nodes are hosted in Frankfurt or Ireland regions to ensure data remains within the European Economic Area and aligns with GDPR requirements.

**United States** Deployments in N Virginia and Oregon support domestic privacy frameworks such as CCPA and CPRA.

**Australia** The Sydney region is used to meet APRA and CPS 234 obligations, particularly for regulated financial environments.

### Cryptographic Enforcement Layer 

Regional isolation is reinforced with strong encryption controls:

- Data at rest is encrypted using **AES 256**
- Data in transit is protected using **TLS 1.2 or TLS 1.3**

This ensures sovereignty is enforced both geographically and cryptographically.

Centralized Orchestration Through a Master Console 
---------------------------------------------------

### The Role of Central Control in a Decentralized System 

While execution and data storage are decentralized, governance and visibility are centralized. This balance allows organizations to scale without losing control.

### Manager of Managers View 

A master level console aggregates real time telemetry from all autonomous nodes. Administrators gain a unified operational view across the entire global fleet without directly managing each shard individually.

### Centralized License Governance 

License consumption across Standard, Enterprise, and Ultimate tiers is tracked globally. This prevents over provisioning and enables accurate operational cost management.

Cross Region Governance and Policy Consistency 
-----------------------------------------------

### Global Compliance Visibility 

Administrators can execute compliance audits that span all shards, allowing security posture to be evaluated across hundreds of thousands of devices in a single workflow.

### Policy Template Propagation 

Standardized security configurations can be deployed simultaneously across all nodes using pre configured policy templates. This ensures consistent enforcement of corporate or regulatory baselines without manual per node configuration.

Architectural Takeaway 
-----------------------

This architecture deliberately separates responsibilities:

- **Regional shards** handle performance, isolation, and regulatory compliance
- **Central orchestration** delivers visibility, consistency, and governance

The result is a system that scales linearly in size without scaling linearly in complexity, allowing global endpoint environments to grow while remaining stable, compliant, and operable.