How does Node.js, a single-threaded environment, achieve scalability on multi-core systems?

Node.js on multi-core computers

The current popularity of Node.js has raised some doubts that it only targets single processes and Threads are optimized, so how does it scale to multi-core CPUs and multi-CPU servers?

Does single threading really limit performance?

Although single-threaded servers are fast enough, under high load we need to take full advantage of multi-core processors. Parallelizing tasks is a current trend in accelerating applications.

Node.js combined with multi-cores

Node.js does scale to multi-core computers. It is implemented through the following methods:

1. Process splitting

For heavy computing tasks such as image encoding, Node.js can launch child processes or provide support to other worker processes Send message. In this design, one thread manages the flow of events, while N processes perform heavy computational tasks, fully utilizing the other 15 CPUs.

2. Multi-server parallelism

For Web service scaling, multiple Node.js servers should be run on one computer, with one core per server and amortized Request traffic. This provides excellent CPU affinity and scales throughput almost linearly as the number of cores increases.

Use the cluster module to scale web service throughput

Starting with v6.0.X, Node.js directly includes the cluster module, making it easy to set up multiple nodes Worker processes that can listen on a single port. Note that this is different from the older learnboost "cluster" module available via npm.

if (cluster.isMaster) {
  // Fork workers.
  for (var i = 0; i < numCPUs; i++) {
    cluster.fork();
  }
} else {
  http.Server(function(req, res) { ... }).listen(8000);
}

Worker processes will compete to accept new connections, and the process with the least load is more likely to win. This works well and scales throughput well on multi-core machines.

Optimize load balancing

In addition to using the cluster module, the following steps can also help to further improve multi-core load balancing:

• In Nginx or Web proxies such as Apache run Node.js services behind them to perform connection throttling, URL rewriting, serving static content, and proxying other sub-services.
• Restart worker processes regularly. For long-running processes, even small memory leaks can accumulate.
• Set up log collection and monitoring.

The above is the detailed content of How does Node.js, a single-threaded environment, achieve scalability on multi-core systems?. For more information, please follow other related articles on the PHP Chinese website!