Build an Orchestrator in Go (From Scratch)

An orchestrator automates the deployment, scaling, and management of containerized applications, abstracting away machines and operating systems to let developers focus on their code. [1]

Deployment evolved from bare metal servers to virtual machines and finally to containers, each step increasing efficiency and standardization. [1] [2]

The main components are the manager, worker, scheduler, tasks, jobs, and the cluster, each with distinct responsibilities in running and managing workloads. [1]

Tasks are run as Docker containers, with the system programmatically starting, stopping, and inspecting containers to manage application lifecycles. [1] [2] [3]

By running multiple worker nodes, the system can distribute tasks across machines, ensuring continued operation even if some nodes fail. [1] [2]

RESTful APIs are implemented for both worker and manager components, enabling remote task management and status queries. [1] [2]

Worker nodes gather system metrics using the goprocinfo library and expose them via an API endpoint for observability. [1] [2]

The book covers both simple round-robin scheduling and a more sophisticated E-PVM scheduler, discussing their tradeoffs and implementation. [1] [2]

Persistent storage for tasks is implemented using BoltDB, allowing the system to recover task state after failures. [1] [2]

A command-line interface (CLI) is built using the Cobra framework, supporting commands to start components, manage tasks, and query system state. [1] [2]