How CDN and Global Load Balancing Supercharge Web Performance

CDN Overview

CDN (Content Delivery Network) is an intelligent virtual network built on existing Internet infrastructure. By deploying edge servers worldwide and using load balancing, content distribution, and scheduling, it delivers content to users from the nearest server, reducing congestion, improving response speed and hit rate.

In short, data is placed on servers in various locations, and load balancing directs users to the nearest server.

CDN Principle

The basic principle uses widespread caching servers distributed in regions where users access. When a user requests a website, global load balancing directs the request to the nearest healthy cache server, which responds directly.

Global load balancing enables sites with servers in multiple regions to be accessed via a single IP or domain, routing users to the nearest server for fastest speed.

CDN aims to avoid network bottlenecks, placing node servers to form an intelligent virtual layer that redirects requests based on traffic, node load, distance, and response time, thereby improving speed and stability.

The goal is to let users obtain needed content locally, alleviating Internet congestion and speeding up website access.

CDN Functions

Save backbone bandwidth and reduce bandwidth demand.

Provide server‑side acceleration, preventing overload from high traffic.

Use web cache to store previously accessed pages and objects locally, reducing backbone usage and improving response time.

Overcome uneven site distribution and lower construction and maintenance costs.

Mitigate “traffic storms” and enhance network stability.

CDN Service Model

CDN is a strategically deployed system comprising distributed storage, load balancing, request redirection, and content management. Content management and global traffic management are core. By judging user proximity and server load, CDN efficiently serves requests. The cache server (surrogate) sits at the edge, typically one hop from the user.

The surrogate mirrors the origin server, allowing CDN providers to deliver optimal experience to end users who cannot tolerate latency.

Global Load Balancing

Global Server Load Balancing (GSLB) balances load across server groups in different geographic locations, enabling users worldwide to access the nearest server via a single IP or domain.

Server Group Selection

The core of GSLB is selecting the appropriate server group for a client, based on proximity and load.

Proximity considers physical distance; methods include static IP mapping or dynamic detection (e.g., hop count).

Load considers server‑group load, such as maximum connections, average response time, and service quality.

Common GSLB implementations: DNS round‑robin, HTTP redirection, IP spoofing (triangular routing).

DNS‑based GSLB

Users resolve domain names via DNS; a GSLB device can replace the DNS server to complete resolution, directing users to the optimal server.

Advantages: simple implementation, low cost.

Disadvantages: decisions based on the DNS resolver’s IP may misrepresent the user’s actual location, leading to inaccurate routing.

HTTP‑Redirect GSLB

To overcome DNS inaccuracies, HTTP‑redirect GSLB sends an HTTP 302 redirect to the chosen server, allowing the real client IP to be used for selection.

Advantages: obtains true client IP, improving accuracy.

Disadvantages: works only for HTTP traffic.

IP‑Spoofing GSLB

IP‑spoofing (triangular routing) works for all protocols. The GSLB device’s IP is advertised in DNS; it forwards the request to the selected server, which replies with the GSLB’s IP as source, making the client unaware of the intermediate hop.

While supporting multiple protocols, it adds extra hops and latency, so it is usually a supplement to HTTP‑redirect.

Server Group Selection Strategies

Geographic or custom regions: map IP prefixes to regions and assign users accordingly.

IP address weighting: assign traffic weights to each IP in DNS responses.

Round‑Trip Time (RTT): choose the server with the smallest RTT, using active or passive measurement.

Active RTT measures via DNS queries or ICMP, which may be blocked and adds traffic. Passive RTT measures the time from SYN to ACK, reflecting actual user experience without extra traffic.