Node.js Worker Threads: Offloading CPU‑Heavy Work to Keep the Event Loop Free

CPU intensive tasks block the Node.js event loop and make an application unresponsive. When a long running calculation runs on the main thread, incoming requests wait until the task finishes.

Node.js runs on a single thread that handles both synchronous and asynchronous work. Worker threads provide a way to move CPU‑bound work to separate threads, leaving the main thread free to handle I/O and other events.

To use worker threads, import the Worker class from the worker_threads module. Create a worker by pointing to a worker file and passing data via workerData. The main thread listens for messages with parentPort.on('message', ...) and handles errors with an error listener.

In the worker file, retrieve parentPort and workerData, perform the calculation, and send the result back with parentPort.postMessage(result). A simple Fibonacci function illustrates the pattern.

To measure the impact, an Express endpoint was built that computes Fibonacci(35) for each request. Autocannon was used with 1000 requests and 50 concurrent connections. Without workers, the average latency was about 4148 ms and the throughput was roughly 11.6 req/s. The test lasted around 86.7 seconds while CPU usage stayed near 27 %.

After moving the calculation to a worker thread, the average latency dropped to about 1646 ms and throughput rose to approximately 29.4 req/s. The test finished in around 34.5 seconds and CPU usage climbed to near 98 %. The higher CPU usage reflects the work being spread across available cores instead of being confined to the main thread.

Using worker threads adds memory overhead because each worker gets its own V8 heap and event loop. Debugging can be more complex due to multiple threads, and transferring large data between threads requires copying unless shared memory is used. Therefore, worker threads are best suited for truly CPU‑bound tasks such as image processing, data compression, or cryptographic operations, while I/O‑bound work should stay on the main thread or use async APIs.

In summary, worker threads let Node.js applications stay responsive under heavy computational loads by offloading work to separate threads. The trade‑off is increased resource usage and code complexity, but for CPU‑intensive workloads the benefit in latency and throughput often justifies the cost.