The Route To High-Availability Networking

May 1, 2000
Because of society's growing dependence on data networking for day-to-day operations, network-equipment providers are under increasing pressure to deliver 100% uptime on the network and backbone. This need also is starting to take a toll on networks...

Because of society's growing dependence on data networking for day-to-day operations, network-equipment providers are under increasing pressure to deliver 100% uptime on the network and backbone. This need also is starting to take a toll on networks of embedded devices, like those that might be found in factory-floor operations and medical devices.

Network equipment, such as hubs and switches, typically runs one or more processors, as well as an RTOS. That operating system provides management of computing resources and the ability to run programs that route or switch TCP/IP packets. These types of systems typically demand high availability and fault tolerance rather than rapid reconfiguration or upgrade.

One network system vendor, Performance Technologies, has implemented a means of hot swapping separate and identical switching modules. These modules use a proprietary means of mirroring switching information. That way, when one fails, the other takes on the entire load with no data or performance loss.

For embedded networking, a proprietary hot-swap approach is less likely to meet requirements because network administrators and repair parts may not be available. Fortunately, Performance Technologies' approach is easily adaptable to CompactPCI and the PICMG 2.1 hot-swap specification. The company is using designs like this to offer hot-swappable networking for embedded applications. With the rapid growth of embedded-networking opportunities, this is likely to be only the first of many implementations.

Sponsored Recommendations

Comments

To join the conversation, and become an exclusive member of Electronic Design, create an account today!