Redundant Web Hosting
Jun 22, 2004

Outside the World Wide Web, redundancy is usually an undesirable trait. No one wants to be accused of being redundant—of saying or writing what has already been said or written. And so we may associate redundancy with being a poor listener or being boring. But in terms of the Internet, redundancy is a technical term that refers to a crucial feature of good web hosting. Learn more about the specialized meaning of redundant.

Redundancy in the World of Web Hosting

Something that is redundant is a duplication. When you're talking about crucial data for a web customer, individual or business, redundancy helps protect the data. Should something go wrong with one copy of the data, there's a redundant (back-up) copy available. Redundancy is available not only for higher priced dedicated web hosting, but also for customers who purchase shared hosting contracts. Regardless of your web hosting package, redundancy is a feature that helps keep your website live and your data safe.

Redundancy provides a stopgap when the system develops a flaw or failure. It contributes to a system's "fault-tolerance"—its ability to continue proper operation in the event of a system failure or one or multiple component faults. The potential failure of a system can be addressed in several ways. Failover is the ability of a system to automatically rebound after a fault or failure, using a redundant server, system, or network to substitute for the one in which the error took place. Switchover is the manual counterpart to failover, in which a person switches to a redundant system. Switchover is the method used to take a server, system, or network offline for system maintenance or upgrade.

Data Redundancy

There are several types of redundancy. Data redundancy (which is not data reliability, though it has been called that), is a term used in computer data storage to describe a property of disk arrays. With a RAID (Redundant Arrays of Independent Disks) system, the entire data set is duplicated or an error-correcting code is stored beside it or, in some cases, on another hard drive. This results in the reduction of available disk space.

More About Redundancy

But redundancy can be far more extensive than this. It is also possible to have redundant servers, redundant power supplies on separate power circuits, redundant networks using this article has all rights reserved and is copyright by 100 Best multiple carriers, and redundant datacenters located at a distance. This much redundancy is likely to be both costly and unnecessary to most enterprises, but it is available for those who require it.

Let's look, for a moment, at what a redundant server set up might entail. A system with load balancing—i.e., when work that could theoretically be done on one server is spread across multiple resources to increase reliability and speed might involve a primary server and a secondary server both connected to a database server, with net traffic arriving through the primary server only, and the two servers connected by a "Heartbeat" cable. The failure of the pulse/heartbeat from the main server to the secondary server signals the secondary server to take over, accomplishing failover. With the primary and secondary server in constant contact, keeping both data sets up-to-date, the system has benefits not afforded by a system with a back-up record that must be recovered and launched for each hosting client.

 

Related Web Tools:
Using a Dedicated IP Address