And why do I need one?

Let's consider a small office that has 3 PCs. While each one may have a different set of installed software it's probable that at least one application is common, such as Microsoft Office.

Chances are that from time to time documents created one one of the PCs may need to be accessible from one of the others. There are several different ways that this may be accomplished:

1. Copy the document to a 'floppy' disk
2. 'Burn' it to a CD or DVD
3. Use a remote access program to transfer it over a network
4. Create a shared location and place the document there

Obviously options 3 and 4 require that the PCs be connected together in a network, but with the importance of the Internet and email that is very likely to be the case.

Going further we can easily believe that all 3 PCs may need to share data amongst themselves. This could be accomplished by creating shared locations on each PC and allow the others to access those locations. With a small network this would seem to be a reasonable configuration. When the user on one PC needs access to a shared document there are only a few locations that need to be checked - the local PC and the shared locations on the other 2 PCs.

So what happens when you increase the number by adding 1 or 2 or more new PCs?

Is it still reasonable to provide shared locations on each one ? At what point can we say that the number of possible locations for a shared document has become unmanageable?

What if we want to restrict access to some documents while allowing all users to have full access to others ? Will we need to add more shared locations on each PC ?

Enter the server.

We install one extra computer and create one or more shared locations on that computer. The main role that this one has is to provide a central location for all your documents. By creating several shares with different access permissions we can protect critical documents from unauthorised access. This is achieved by creating accounts on the server and forcing users to login with their accounts before they can access the network resources - this is the secondary role of the server.

Further, by centralising the data we can install some kind of backup device such as the RDX QuikStor on the server and schedule a daily procedure to copy all the documents to removable media to protect against accidental loss.

Why have a separate server if it's just used for storing and protecting data ?

Server software is optimised for the provision of background services while workstation software such as Windows XP or Vista is optimised for the foreground applications. Also the hardware used for a server is more robust than that generally used for the client PCs. Disc drives can be configured in what are known as redundant arrays such that the server can continue to operate even if one of the drives fails.

There are many other services that can be installed such as Terminal Services that allow multiple users to connect remotely and run their programs directly from the server, Database services that can provide a central location for client program storage, web services, email services, etc

Conclusion.

For all but a very small network the inclusion of a server can simplify administration and make it easier to manage both data storage and protection.