Load balancing is the even distribution of computer processing and communication activities so that a server is not overwhelmed. Load balancing is especially important for networks where it is difficult to predict the number of requests that will be issued to a server.

Let Rackspace make the difference with reliable managed hosting solutions backed by Fanatical Support.™ 100% Network Uptime Money-Back Guar., 1-hour hardware replacement, 24/7 LIVE, MS and LNX certified technical support.

Busy Web sites typically employ two or more Web servers in a load-balancing scheme. If one server starts to get swamped, requests are forwarded to another server with more capacity. Load balancing is therefore a service that is employed on more sophisticated site deployments.

Most service providers that offer managed hosting will provide load balancing services to their clients if they experience constantly heavy traffic on their Web sites. Multiplexing traffic amongst several servers permits better site availability. Load balancing therefore distributes traffic efficiently among network servers so that no individual server is overburdened.

In the early days of the technology, many load balancing solutions where implemented through customized domain name server configurations. This approach called “DNS round robin,” rotated incoming requests from Web browsers to multiple Web servers. The result was that all requests to the particular Web site were evenly distributed among all of the machines in the cluster.

The advantage of this system was that it was inexpensive, as a system administrator only needed to make a few changes in order to implement it. Its major disadvantages however was that DNS round robin did not support server affinity or high availability.

Server affinity is a load-balancing system's ability to manage a user's requests, either to a specific server or any server, depending on whether session information is maintained on the server or at an underlying, database level. Without server affinity, DNS round robin could not by itself exercise control over incoming traffic without the aid of cookies, hidden fields or URL rewriting.

Due to its very nature of not being able to exercise this type of control, DNS round robin could not natively handle browser caching, since incoming requests were randomized. DNS round robin also had tremendous difficulties dealing with high availability, since the DNS system requires a large amount of time of propagate information and cache it throughout the network. As a result, DNS round robin is a more antiquated approach toward managing server load. Load balancing, however, is much more sophisticated today, because it is mainly implemented through appliances such as smart switches.

Hardware load balancers solve many of the problems faced by the round robin software solution through virtual server addressing. A load balancer shows a single, virtual server address to the outside world, which maps to the addresses of each server within the cluster. When a request comes to the load balancer, it rewrites the request's header to point to other machines in the cluster. If a machine is removed from the cluster, the request does not run the risk of hitting a dead server, since all of the machines in the cluster appear to have the same IP address. This address remains the same even if a node in the cluster is down. Moreover, cached DNS entries around the Internet are not a problem. When a response is returned, the client sees it coming from the hardware load balancer machine. In other words, the client is dealing with a single machine, the hardware load balancer.

Because one machine is dealing with the management of an entire cluster, server affinity and high availability is actually integrated into the hardware load balancer solution. Indeed, with the hardware solution, a service provider can offer fail-over services, cluster management, traffic routing and geographic load balancing across multiple sites and points of presence.

Hardware also is a much faster traffic routing solution because it is solid-state device that more efficiently processes requests than software on a server deployment can.