Pulse is under active development. You may encounter bugs. If you do, please contact us.Contact support

Pulse Documentation

Pulse is a mobile-first infrastructure monitoring platform. This documentation explains how the system works, how the agent operates, and how telemetry flows from your servers to the Pulse mobile application.

Platform Overview

Pulse consists of a lightweight monitoring agent, a centralized backend, and a mobile application. Together, these components provide continuous visibility into server health and Docker workloads.

  • Pulse Agent: Runs on Linux servers and collects metrics.
  • Pulse Core: Authenticates agents, processes telemetry, and stores time-series data.
  • Pulse App: Mobile dashboard for real-time and historical insights.

System Architecture

The Pulse architecture is designed around secure outbound communication. Agents periodically collect system and container metrics and transmit them to the backend API over HTTPS.

Agent Layer

Runs on each server. Responsible for host inspection, container discovery, metric snapshots, and heartbeat reporting.

Backend Layer

Validates API keys, persists telemetry, aggregates metrics, triggers events, and exposes APIs to mobile and web clients.

Client Layer

Mobile application for server visibility, container monitoring, notifications, and remote interactions.

Monitoring Agent

The Pulse Agent is a Go-based service designed for long-running execution. It performs scheduled inspections, gathers Docker statistics, generates structured telemetry snapshots, and maintains server presence through heartbeats.

  • Collects host CPU, memory, disk, uptime, and OS metadata
  • Discovers running containers and captures runtime statistics
  • Associates servers with API keys and environments
  • Maintains a heartbeat to track server availability
  • Operates without exposing any inbound network ports

Metrics & Telemetry

Pulse captures both real-time snapshots and aggregated metrics. These data streams enable current health views, historical analysis, and alerting.

System Metrics

CPU utilization, memory usage, disk consumption, uptime, architecture, hostname, and operating system information.

Container Metrics

Container identity, image, runtime state, CPU usage, memory consumption, limits, and network activity.

Agent Lifecycle

Each Pulse Agent follows a predictable lifecycle designed for reliability and continuous operation.

  • Startup and environment validation
  • API key authentication and server registration
  • Periodic metric collection
  • Telemetry transmission to backend
  • Heartbeat updates and version reporting
  • Graceful recovery from network interruptions

Security Model

Pulse is built with a minimal-access security model. Agents authenticate using scoped API keys and communicate only through outbound HTTPS requests.

  • API key–based authentication
  • HTTPS-only transport
  • Read-only Docker socket access
  • No inbound open ports
  • No shell execution or remote code paths
  • Isolated server identity and environment tagging