39

  • Did I understand correctly: If I use cluster package, does it mean that a new node instance is created for each created worker?

  • What is the difference between cluster and worker_threads packages?

Oliver Sieweke
  • 1,812
  • 2
  • 14
  • 23
Денис Райгородский
  • 493
  • 1
  • 4
  • 6

1 Answers1

70

Effectively what you are differing is process based vs thread based. Threads share memory (e.g. SharedArrayBuffer) whereas processes don't. Essentially they are the same thing categorically.

cluster

  • One process is launched on each CPU and can communicate via IPC.
  • Each process has it's own memory with it's own Node (v8) instance. Creating tons of them may create memory issues.
  • Great for spawning many HTTP servers that share the same port b/c the master main process will multiplex the requests to the child processes.

worker threads

  • One process total
  • Creates multiple threads with each thread having one Node instance (one event loop, one JS engine). Most Node API's are available to each thread except a few. So essentially Node is embedding itself and creating a new thread.
  • Shares memory with other threads (e.g. SharedArrayBuffer)
  • Great for CPU intensive tasks like processing data or accessing the file system. Because NodeJS is single threaded, synchronous tasks can be made more efficient with workers
snewcomer
  • 2,020
  • 1
  • 19
  • 22
  • For worker_threads: One process total. Does that mean spread across multiple V8 instances ? – joedotnot Jan 04 '21 at 01:22
  • 7
    Each worker thread will have it's own V8 instance and Event Loop. This is how you gain parallelism. They call it a V8 isolate. Effectively, this is an independent instance of the V8 runtime that has its own JS heap and a microtask queue. Each Node.js worker thread can thus run it's code in isolation without sharing their heaps. Communication; however, occurs through a message channel between child and parent. – snewcomer Jan 05 '21 at 14:21
  • 1
    @snewcomer Wouldn't worker threads involve in deadlock situations as they share memory? If yes, then wouldn't it slow down the CPU performance? – Sarim Javaid Khan May 02 '22 at 00:47
  • Why do you strikethrough master ? – CDT Jul 16 '22 at 11:34
  • 1
    @SarimJavaidKhan They do not share memory. data passed over channels is copied, but this is still way more efficient than JSON stringify/parse for IPC. – Daniel Ennis - Aikar Nov 03 '22 at 05:15