Boost Your NGINX Performance: 8 Proven Tuning Tips

What is NGINX?

NGINX is a modern, open‑source web server that can also act as a media streamer, mail proxy, reverse proxy, load balancer, and cache server.

It is used by many large companies such as VMware, IBM, Cisco, Apple, Microsoft, LinkedIn, Netflix, Facebook, and Twitter because of its high performance and ease of configuration.

Prerequisites for Tuning NGINX on Linux

Before modifying NGINX, you need to:

Deploy and configure an NGINX server on Linux (e.g., follow the guide at https://www.linuxmi.com/debian-11-nginx.html).

Have a basic understanding of NGINX and its configuration files.

Once these requirements are met, you can start optimizing NGINX.

1. Configure Worker Processes

NGINX consists of a master process and multiple worker processes. By default, the number of workers is set to auto, which equals the number of CPU cores.

linuxmi@linuxmi:~/www.linuxmi.com$ grep processor /proc/cpuinfo | wc -l

On the author's test machine there is only one core.

To increase the number of workers, edit /etc/nginx/nginx.conf with a text editor such as nano:

linuxmi@linuxmi:~/www.linuxmi.com$ nano /etc/nginx/nginx.conf

Set the worker_processes directive to the maximum number of available CPU cores.

2. Configure worker_connections

The worker_connections parameter defines the maximum number of simultaneous TCP connections each worker can handle. The default is 512, but modern systems often support higher values. linuxmi@linuxmi:~/www.linuxmi.com$ ulimit -n Adjust worker_connections in the events block of nginx.conf to match the supported limit.

3. Enable GZIP Compression

GZIP reduces bandwidth usage and speeds up page loads on slow connections. Add the following lines to the server block:

gzip on;</code>
<code>gzip_vary on;</code>
<code>gzip_min_length 10240;</code>
<code>gzip_proxied expired no-cache no-store private auth;</code>
<code>gzip_types text/plain text/css text/xml text/javascript application/x-javascript application/xml;</code>
<code>gzip_disable "MSIE [1-6]\.";

4. Reduce Timeout Values

Lowering timeout values reduces CPU and network overhead caused by opening and closing connections. Add the following directives to the http block:

client_body_timeout 12;</code>
<code>client_header_timeout 12;</code>
<code>keepalive_timeout 15;</code>
<code>send_timeout 10;

5. Adjust Buffer Sizes

Too‑small buffers cause NGINX to write temporary files, increasing I/O. Set these buffer parameters in the http block:

client_body_buffer_size 10K;</code>
<code>client_header_buffer_size 1k;</code>
<code>client_max_body_size 8m;</code>
<code>large_client_header_buffers 4 4k;

6. Disable Access Log or Enable Log Buffering

Logging every request consumes disk space and CPU. To disable logging, add: access_log off; If you need logs, enable buffering to write logs in batches:

access_log /var/log/nginx/access.log main buffer=16k;

7. Cache Static Content

Cache static files (images, CSS, JS) to reduce bandwidth and improve response time. Add the following location block to your virtual host configuration:

location ~* \.(jpg|jpeg|png|gif|ico|css|js)$ {</code>
<code>    expires 100d;</code>
<code>}

This caches the files for 100 days from the last access.

8. Enable Open File Cache

Open file caching stores frequently accessed file descriptors, reducing disk I/O. Add these directives to the http block:

open_file_cache max=1024 inactive=10s;</code>
<code>open_file_cache_valid 60s;</code>
<code>open_file_cache_min_uses 2;</code>
<code>open_file_cache_errors on;

When modifying the configuration, change one setting at a time and test it before proceeding to the next. This systematic approach ensures stable performance improvements.