Toolkit.
Simulator · Study Hall

Goroutine Scheduler

Visualize how Go schedules goroutines across OS threads using the GMP model.

Interactive
Push on the parts yourself.

The goroutine scheduler workspace from the original build will sit here — same logic, same controls, restyled for Study Hall. Prose below covers what you'll be able to do.

Go uses an M:N scheduling model where M goroutines run on N OS threads. Understand the GMP model (Goroutine, Machine, Processor) and see how work stealing keeps all processors busy.

The GMP model

G (Goroutine) is a unit of work. M (Machine) is an OS thread. P (Processor) is a context that runs goroutines on a machine. Each P has a local run queue; global queue handles overflow.

Work stealing

When a P local queue is empty, it steals half the work from another P queue. This balances load across processors and maximizes CPU utilization.

Good for

  • Understanding Go concurrency
  • Optimizing Go applications
  • Learning runtime internals
  • Debugging performance issues

Questions people ask

How many threads does a Go program use?

GOMAXPROCS sets the number of Ps (logical processors). The runtime may create more Ms (threads) for blocking operations like syscalls or CGO calls.

What triggers goroutine scheduling?

Function calls, channel operations, system calls, and explicit yields (runtime.Gosched) are scheduling points where the runtime may switch goroutines.