Concurrency vs Parallelism in Node.js: Understanding the Event Loop & Worker Threads
summary
Node can handle thousands of concurrent connections on one thread but stalls on CPU-heavy work, so use non-blocking I/O + event loop for API and real-time feeds and move heavy processing to parallel options: child_process (spawn/fork/exec/execFile), cluster, or worker_threads. Microtask ordering is concrete: promises before setTimeout; worker_threads share memory via SharedArrayBuffer instead of separate heaps. A local test on a 12-core MacBook M3 shows 12 cluster workers each with unique PIDs.
AI-written summary · about
chapters (9)
- 0:00Concurrency vs parallelism
- 2:28Node.js event loop
- 4:55Microtasks vs macrotasks
- 7:25Concurrency with event loop
- 8:10Parallelism with child processes
- 9:13Scaling with cluster module
- 10:14Worker threads for CPU tasks
- 11:19Key differences summary
- 11:40Q&A on worker threads
related talks
35mNodeJS: from single thread to worker threads
33mNode.js: The Road to Workers
22mScaling Node.js and unblocking the Event Loop
26mA Crash Course on Worker Threads
30mParallel Universes: Node.js
27mA Crash Course On Worker Threads
23mDecomposition of the Main Thread in Node.js to increase throughput
34mNode.js: The Road to Workers