Load Balancing Basics

Even regular internet traffic demands a great deal of resources from a web server.  Mere routine traffic is enough to make the hardware perform sluggishly and eventually fail once it is overloaded.  One factor that has proven to effectively mitigate these issues is a technology known as load balancing.

What is Load Balancing?

Load balancing is a process that allows internet traffic to be evenly distributed across two or more servers.  It essentially covers resources such as CPUs, hard drives and network links, resulting in optimal resource utilization, maximum throughput, and faster loading times.  Load balancing is particularly useful in cluster configurations and busy networks where it tends to be difficult to determine or predict the number of requests that will be made to a server.

Load Balancing in Action

Load balancing is a relatively straightforward technology.  When a request for a web page is made, it is sent to the load balancer, which subsequently forwards the request to one of the servers.  From there, the server sends a response back the load balancer, which then delivers the output to the end-user.

In the event that one machine gets overloaded, the incoming requests are forwarded to another server.  Load balancing is able to dramatically reduce downtime and increase availability.  Even if one server should fail, another is there to take on the load.  In a GSLB (Global Server Load Balancing) configuration, the load is geographically distributed to scattered servers based on criteria such as proximity, load or health.

We demand a lot from servers, and load balancing is one of numerous methods that ensure they continue to thrive.