Device Management Service (DMS)
The Device Management Service (DMS) is a core component of the NuNet ecosystem that enables users to connect, configure, and manage their computing devices as part of NuNet’s decentralized network.
Think of DMS as the “control center” that allows your machine, whether a laptop, desktop, server, or edge device, to become part of the Global Economy of Decentralized Computing.
Overview
In simple terms, the DMS acts as the bridge between individual compute nodes (devices) and the NuNet network orchestration layer.
It:
- Detects and registers your device on the NuNet network.
- Manages its identity, configuration, and communication with the network.
- Enables computational workloads to be allocated to your device securely.
The DMS makes it possible for anyone, from individuals to organizations, to offer spare compute resources to the NuNet network or run applications on distributed devices.
Core Responsibilities
The Device Management Service handles several critical functions to ensure seamless participation in the NuNet ecosystem:
1. Device Registration and Identity
Each device connected to NuNet is uniquely identified.
When you install and launch DMS, it:
- Generates a unique cryptographic identity (based on a keypair).
- Registers this device identity with the NuNet network.
- Allows your device to securely authenticate with other nodes and services.
This ensures all interactions are trusted, verifiable, and decentralized, without central user accounts or passwords.
2. Resource Discovery and Monitoring
DMS continuously monitors the resources available on your machine, including:
- CPU cores and utilization
- Memory (RAM)
- Disk space
- GPU availability
- Network connectivity
These metrics are reported to NuNet’s orchestration layer, which uses them to match decentralized workloads (called allocations) with devices capable of executing them.
This allows DMS to advertise your device’s capabilities to the network while ensuring privacy and security.
3. Secure Workload Management
Once your device is part of the NuNet network, DMS can receive and manage tasks or workloads assigned to it.
For example:
- A DMS instance on a laptop might run a containerized machine learning inference task.
- A DMS instance on a server might process data for a distributed computation pipeline.
- A DMS instance on an edge device might host a lightweight web service.
Each workload runs inside secure, sandboxed environments, typically Docker containers, ensuring isolation between your system and the task.
4. Decentralised Peer to Peer Orchestration
The DMS enables orchestration of computational workfloads over nodes in the Nunet Network.
5. Contracts and Payments
DMS allows peers in the network to cryptographically sign contracts with the agreed terms of engagement. And then compensate the resource providers based on these terms.
Architecture Overview
At a high level, DMS consists of several internal modules:
| Module | Description |
|---|---|
| Core Service | The main process that initializes and maintains DMS operations. |
| Resource Monitor | Collects system metrics (CPU, RAM, disk, GPU). |
| Network | Handles peer discovery, communication, and connection management. |
| Executor | Runs tasks in isolated environments (e.g., Docker containers). |
| Security Layer | Manages key generation, signing, and encrypted communication. |
| Tokenomics Layer | Manages contract signing, verification and payments. |
All modules work together to provide a robust, secure, and user-friendly experience for both compute providers and consumers.
Security and Privacy
NuNet places a strong emphasis on security, privacy, and decentralization.
DMS follows several principles to ensure these goals:
- Local Control: All workloads and configurations are executed on your device under your control.
- Cryptographic Authentication: Every message is signed with your device’s keypair and authenticated at the receiving end.
- Isolation: Tasks run in isolated containers, protecting your system.
- No Central Authority: Devices authenticate and communicate peer-to-peer without relying on a central server.
This means you retain full ownership of your device, data, and compute power — while still participating in a global, collaborative compute network.