Fault Tolerance Errors in Network

Fault Tolerance

Errors in network can result in serious financial problems for a company. In addition to the monetary loss of more than $50,000 per hour, it can cause frustration to network administrators and customer alike. The example of airline network may help explain this better. When a ticket reservation system online suddenly breaks down, it means hundreds of customers from around the globe can no longer reserve their tickets online. This means they will usually move to some other service providing tickets and this would cause the company millions of dollars in revenue.

This has led to an increased demand for products such as magnetic tape backup, recording systems, UPS and fault tolerance systems that would help in protecting their network systems against data loss. In most companies, it is the job of the network manager to use some kind of backup system to regularly create archival files of data. Apart from these, they also use UPS, which helps in keeping the system operational in case of power failures. These are some of the standard technologies implemented to protect data against potential threats.

While these technologies are standard, there are some specialized fault-tolerant products such as Novell's System Fault Tolerant (SFT) NetWare are not that well understood and hence not used that often. Fault tolerant devices are also called mirroring products since they work on redundancy by providing making identical copies of data and program files. The fault tolerant devices make use of the server and the hard disk to mirror files and each of these files have the exact same formatting and content as the original copy.

Fault tolerant technologies help in providing security to a large extent. This kind of security is more protective than simple back up or data recovery system. A fault tolerant network is not the same thing as a fault resilient system. As Brian Walsh explains:

Fault tolerance means the system will not fail because any one component fails. The system also should provide recovery from multiple failures. Components are often over engineered or purposely underutilized to ensure that while performance may be affected during an outage the system will perform within predictable, acceptable bounds... Fault resilience usually indicates that at least one of the modules within a component, say, a power supply in a hub, is backed up with a spare. Not all modules within that component are necessarily redundant, however. The hub may have two power supplies but only one CPU. Performance of the system during an outage is therefore undefined. One fault-resilient component does not make the entire system fault tolerant."

Some of the key methods by which a network is made fault tolerant include identical disk mirroring, server mirroring system or disk duplexing. In the first method, two hard disks are used to create identical copies of data and program files. In server mirroring method, two servers are used for the same purpose while in disk duplexing, two disk operators are used instead of one.

Some important available products for fault tolerance include:

Ethernet's complete fault tolerant system. This is called Telemecanique® Connexium (TM) IP67 Ethernet switch. Along with this Ethernet also offers a wide range of products to make the system more resilient. A website indicates: "The Connexium family of Ethernet industrial hubs, switches, bridges and transceivers enable a complete Ethernet solution that supports a high level of resilience. Their scalable redundant features-from single to double-ring structures-make it easy to build fault-tolerant networks that fit specific environments."

NEC Fault tolerance server provides solution in a single box. These are two servers designed to provide non-stop power supply in order to protect data and avert potential risks to businesses in case of power failure. The two products available by NEC are the NEC Express5800/320Lb Fault Tolerant Servers, and the NEC Disaster Recovery solution. Novell similarly offers a large variety of fault tolerance solution including the SFT series, MTO and stand-by servers.

Windows is not far behind in this field. Windows Servers series is publicized as the most cost efficient fault tolerance solutions. Defining fault tolerant servers offered by Windows, the company states that these servers are those that provide redundancy across all hardware components. This means that if one component fails to work, the secondary component would take over without any time loss. Some of the enhanced features of Windows Server 2003 include: