Name that device! Design your naming model properly

Manageable, scalable, and efficient networks share a common theme: a well-designed naming model. Warren Heaton serves up a few pointers for those seeking to strengthen their naming conventions.

What’s in a name? Well, when it comes to designing manageable, scalable, and efficient networks, the device name can contain a wealth of information.

Names are assigned to many types of devices in today’s modern networks—workstations, printers, servers, switches, and routers. A well-designed naming model should allow users to access these devices by name rather than network address. Most network protocols require a device to have a network address, and the end user’s system must map this address to a name. The methods used to map the network address to a network name can vary from static HOST files to Dynamic DNS. Dynamic methods of name resolution usually are the best solution.

Being obvious is a good thing
For a naming scheme to be manageable, network names should be short and meaningful. Avoid using names like Olympus or Catharsis. While these names may have inter-stellar meaning, they give the network administrator very little information about the device. Instead, use names that give information about the type of device and the device location.

For example, for a router located at a branch office in Louisville, KY, consider using something like sdfbranchrtr. In this example, sdf is Louisville’s airport code, branch describes the location, and rtr tells you that this device is a router.

Many network designers use airport codes as a geographical prefix and abbreviations, such as rtr for routers or svr for servers, as a defining suffix. The information you place between the prefix and the suffix is up to you, but you should follow some general guidelines.
When working in a high-security environment, using names that are very descriptive may present a security issue. As with all of your network design decisions, security must be weighed against manageability and performance.
While many network protocols and applications allow for spaces in network names, many legacy applications and some protocols do not; therefore it is best to avoid using spaces. Additionally, do not use unusual characters such as % or !, as these characters may have special meaning in some applications or protocols and can cause unpredictable results. Furthermore, try to avoid using case-sensitive names. Case-sensitive names are more difficult to type, and it is harder to remember which characters are capitalized and which are not. Lastly, if you are working in an older DOS-based environment, you must consider the old eight-character naming rule.

Warren Heaton CCDA, CCNA, MCSE+I is the Cisco Program Manager for A Technological Advantage in Louisville, KY.

If you'd like to share your opinion, please post a comment at the bottom of this page.