Green Threads

Green threads are threads that are scheduled by a virtual machine (VM). In contrast, typical threads are scheduled by the underlying operating system. Green threads emulate multithreaded environments without relying on any native OS capabilities, and they are managed in user space instead of kernel space, enabling them to work in environments that do not have native thread support.

Native threads can switch between threads pre-emptively, switching control from a running thread to a non-running thread at any time. Green threads only switch when control is explicitly given up by a thread (Thread.yield(), Object.wait(), etc.) or a thread performs a blocking operation (read(), etc.). Whether a blocking call causes yielding depends on the virtual machine implementation.

On multi-CPU machines, native threads can run more than one thread simultaneously by assigning different threads to different CPUs. Green threads run on only one CPU, because every green thread runs on the same operating system thread. This means that a green thread that blocks will block every other green thread as well.

Green threads can be started much faster on some VMs. They significantly outperform Linux native threads on thread activation and synchronization. Native threads have better performance on I/O and context switching operations, however.

Green threads are a good choice over native threads if the platform does not provide support for threading. Another reason is that if your code is not thread-safe or if you need to spawn many threads and often since this is cheaper with green threads.

The Erlang virtual machine has what might be called 'green processes' - they are like operating system processes (they do not share state like threads do) but are implemented within the Erlang Run Time System (erts). These are sometimes erroneously cited as green threads.

Summary

Green threads are rarely used nowadays. They provide little use and have their drawbacks.

When to use

• The underlying platform does not provide support for threads.
• We need to spawn many threads or often.
• Our code is not thread-safe.