There's no such thing as a perfect naming scheme for computers on a Windows network.
Every computer on a network requires a unique name. This can be the source of frustration for systems administrators.
Over the years, I've tried many techniques for naming hosts on Windows networks. Some worked out well; others were disasters. I'd like to share some of what I've learned about naming network hosts.
In the early 1990s, I was a network administrator for a big insurance firm that was switching to Windows from a NetWare network. Even back then, every workstation had to have a unique computer name. This being my first real experience with Windows networking, I decided that the easiest way to ensure that each workstation had a unique name was to name each workstation after the employee who'd be using it.
That may have been the worst idea, IT-wise, I've ever implemented. Why? Users do not keep the same PCs forever. At the time, my employer was growing rapidly, and was hiring new employees right and left. Any time an employee was hired, the company purchased a computer for them.
Since that new computer was usually more powerful than any of the existing systems in the new employee's department, it soon became a tradition for the department manager to take the new PC and pass his old PC off to the department supervisor. The department supervisor would then pass their PC off to the next person in the pecking order. And so on and so on.
This is not a good way to do business. The company's IT department wasted a lot of time and resources with such moves. The purchase of a new PC sometimes meant moving a dozen PCs to ensure that the highest-ranking employees in the department had the latest hardware.
In a corporation with this type of culture, naming a PC after the person who was going to be using it was a huge mistake. But even at a company that doesn't shuffle workstations every week, naming computers after users is still a bad idea. Some employees will always be leaving.
Size of the network is key factor
So how should you name computers? It depends on the size of the network. On networks with fewer than 50 computers, you can get creative with the naming without having to face any real consequences. On several occasions I've built networks for friends who have small businesses. One time I named the computers after the friend's favorite bars; another time I named them after cartoon characters. But fun naming schemes like these aren't practical for larger networks.
Since Microsoft does not provide any real rules for naming hosts on Windows networks, the best advice I can give is to choose a naming scheme that works for your own individual situation.
When I attempted to name hosts on a network based on username, I figured that if I named each PC after the user who used it, I'd always know exactly which PC each NetBIOS name referred to. Today this is still important. Many companies use workstation management software for tasks such as deploying applications, applying service packs or simply monitoring for problems.
I've seen a lot of large networks in which workstation names consist of a hexadecimal number based on a hardware hash or other unique value. Although this technique ensures that host names are unique, it may be the worst method for naming network hosts. Suppose you were using an automated solution to deploy a critical security patch to all your company's workstations, and that after the patch was deployed, you checked the deployment software's logs and found that the patch failed to be deployed on a machine named A2FF68D3. How would you know which PC this name actually refers to?
If you wanted to look at the individual PC, you could always use remote management software to attach to the PC by computer name. But if your corporate security policy forbids the use of remote access, you'd have to find another way of determining which physical PC was having the problem.
Provide clues to PC location This is why I think that the best naming conventions provide clues to a PC's location. For example, basing the naming convention on department, geographic location or job function would allow you to narrow down a PC's location without basing a name on an individual user account. I already mentioned the difficulty in locating a PC that uses a random hexadecimal value as its computer name. But imagine if your workstation management software reported a problem with a PC named "Marketing Dept – Proposal Coordinator." If locating this individual PC became an issue, you could just send someone up to the marketing department and have them ask where to find the proposal coordinator. This naming technique works because employees may come and go, but the job function remains the same (at least in some companies). If you know that the marketing department is always going to have a proposal coordinator, there's no reason why you can't base the computer name on the job title of the computer's user. This technique does not work at every company. What happens if the marketing department has 150 employees who all have the same job title? In a situation like that, you have to get into some kind of numerical naming scheme (proposal coordinator 1, proposal coordinator 2, and so on), including a geographical or a department reference along with the numbers. This will at least help you narrow down the location of the PC.
About the author: Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Windows 2000 Server, Exchange Server and IIS. He has served as CIO for a nationwide chain of hospitals and was once in charge of IT security for Fort Knox. He writes regularly for SearchWinComputing.com and other TechTarget sites.
More information on this topic:
- Tip: Change your network adapter's IP information
- Topics: LAN management
- RSS: Sign up for our RSS feed to receive expert advice every day.