Understanding Nginx: Core Concepts, Features, and Architecture

1. Concept

Nginx (Engine X) is a free, open‑source, high‑performance HTTP server and reverse‑proxy server; it also functions as an IMAP, POP3, and SMTP proxy. It can host websites directly, similar to Tomcat, and solves the C10K problem using an asynchronous, event‑driven architecture.

Although its market share is modest, its high performance and low memory consumption have made it increasingly popular for load‑balancing scenarios, e.g., WordPress and GitHub.

2. Forward Proxy vs. Reverse Proxy

A forward proxy operates on the client side, forwarding client requests to external servers (e.g., GoAgent for bypassing firewalls). A reverse proxy operates on the server side, receiving client requests and dispatching them to backend servers; Nginx is a classic reverse‑proxy.

3. Types of Servers

1) Web servers (e.g., Nginx, Apache, IIS) provide HTTP access. 2) Application servers run application code. 3) Proxy servers mediate client access. 4) Reverse proxies. 5) Backend servers. 6) CDN cache servers, which are reverse‑proxy‑based caching solutions.

4. Nginx Features

Ø Cross‑platform: runs on most Unix‑like systems and has a Windows port. Ø Simple configuration: easy to learn. Ø Non‑blocking, high‑concurrency: supports up to 50,000 concurrent connections in tests, 20‑30k in production, thanks to an epoll‑based event model. Ø No long connections between Nginx and backend web servers. Ø Asynchronous request handling: receives the full client request before forwarding to backends, reducing backend load. Ø Simultaneous response streaming from backends to clients. Ø Low network dependency: works as long as the server is reachable via ping. Ø Built‑in health checks based on status codes and timeouts. Ø Master/worker multi‑process architecture. Ø Low memory usage, cost‑effective, bandwidth‑saving, and highly stable.

5. Basic Functions

Basic HTTP capabilities include serving static files and indexes, caching for reverse proxy and FastCGI, simple load balancing and fault tolerance, modular filters (gzip, byte‑range, chunked, SSI), SSL/TLS with SNI, IMAP/POP3 proxy, and SMTP redirection after authentication.

Additional HTTP features cover name‑based and IP‑based virtual hosts, keep‑alive and pipelining, hot‑reloading configuration without downtime, flexible logging and log rotation, error‑code redirection, and speed limiting.

6. Core Modules

The core of Nginx consists of the CoreModule and EventsModule, plus the third‑party HttpCoreModule which handles HTTP processing. CoreModule controls basic server functions, EventsModule manages connection handling, and HttpCoreModule provides essential HTTP services.

7. Internal Process Model

Nginx primarily uses a multi‑process model (with optional threading). After startup, a master process spawns multiple worker processes. The master manages workers, handling signals and monitoring their status, while workers independently accept and process client connections. Typically the number of workers matches the CPU core count.

8. Request Handling Flow

On startup, Nginx parses its configuration to determine listening IPs and ports, creates listening sockets, and forks worker processes. Workers compete to accept new connections; once a connection is accepted, Nginx creates an

ngx_connection_t

structure, registers read/write event handlers, exchanges data with the client, and finally closes the connection.

9. Real‑World Deployments

Developed in Russia, Nginx has run for years on the Russian portal Rambler Media and is used by over 20% of Russian virtual‑host providers. In China, major sites such as Taobao, Sina Blog, Sina Podcast, NetEase News, 56.com, Discuz!, Douban, and others employ Nginx as a web server or reverse proxy, often together with Tomcat for high‑performance load‑balancing clusters.