What Is a Load Balancer? Definition and How It Works

April 15, 2026·1 min read·167 words

A load balancer distributes incoming traffic across multiple servers to deliver high availability, performance, and scalability. It is essential for web applications, databases, and SaaS platforms. Technologies include L4 (TCP/UDP), L7 (HTTP/HTTPS), and GSLB (global).

Layers

L4 (TCP/UDP): distribution based on IP address/port
L7 (application): distribution based on URL, cookie, header, or content
GSLB (Global Server Load Balancing): multi-datacenter distribution via DNS

Algorithms

Round-robin: sequential distribution
Least connections
Weighted round-robin
IP hash
Response time
URL hash (caching)

Health Checks

TCP connect
HTTP GET with 200 status code validation
Custom scripts
Automatic removal of unavailable backends
Monitoring alerts

Vendors

F5 BIG-IP: enterprise leader
Citrix NetScaler: enterprise solution
HAProxy: powerful open-source solution
NGINX +: commercial web proxy + LB
AWS ALB/NLB, Azure App Gateway, GCP Cloud LB: cloud-native solutions
Kemp LoadMaster: mid-market solution

Additional Features

SSL termination
WAF (Web Application Firewall)
Bot protection
DDoS mitigation
API gateway
Caching

Order from OPTINOC

Load balancers from F5, Citrix, Kemp, and HAProxy. High-availability design + SSL + WAF. Quote within 48 hours.

Frequently Asked Questions

Load balancer ou reverse proxy ?

Un load balancer répartit le trafic (HTTP, TCP, UDP) entre serveurs. Un reverse proxy traite les requêtes (cache, SSL termination, header rewriting) en plus. Beaucoup d'équipements (F5 BIG-IP, NGINX Plus) font les deux.

Hardware ou software ?

Hardware (F5, Citrix ADC, A10) : performances maximales, latence basse, recommandé pour grands débits enterprise. Software (NGINX, HAProxy) : flexibilité, intégration DevOps, gratuit ou abordable. Cloud (AWS ELB, Azure Load Balancer) : standard cloud-native.