Goroutines And Green Threads

What Lightweight Concurrency Means

Operating-system threads are not the only way to run many concurrent tasks. A runtime can multiplex many language-level tasks onto a smaller number of OS threads. Go calls its units goroutines. Erlang and Elixir run lightweight BEAM processes. Java has virtual threads. Crystal has fibers.

The common promise is cheap task creation and blocking or waiting that the runtime can manage. The details differ sharply: BEAM processes have isolated heaps and message passing, Go goroutines share an address space, Java virtual threads keep Java's thread API shape, and fibers may cooperate with an event loop.

Why It Matters

Lightweight concurrency is useful for servers, background workers, pipelines, network clients, actor-style systems, and programs with many mostly-waiting tasks. It can make code read sequentially while still allowing the runtime to schedule lots of work.

It does not remove design limits. A program can still create too much work, hold too much memory, block on native calls, leak tasks, ignore cancellation, or overload downstream systems.

Watch Points

Ask how tasks are scheduled, how blocking calls behave, how cancellation works, how stacks grow, how memory is reclaimed, and whether task-local state can leak across requests. Also ask whether the runtime model encourages shared memory, message passing, or both.

Related Concepts

This concept sits between Threads And Shared Memory, Actor Model And Message Passing, Async Await And Event Loops, and Structured Concurrency.