Client/Server Applications Made Easy

Logical Addressing

To be able to communicate with other computers on a network, a computer needs to be identifiable. Every computer on the network will have its own network adapter that has a unique and permanent physical address. This is what makes each computer identifiable. This address is also used to deliver data to a particular computer. This means that a computer's network adapter has to listen to every single transmission to determine if a particular message is addressed to its physical address.

You can imagine that if you have twenty computers or more, your network will be very slow and inefficient at delivering messages. To reduce network traffic, administrators divide them into smaller networks (called subnets) and then set a hierarchical design so that a message can travel more efficiently to its destination.

TCP/IP provides this subnetting capability through logical addressing. In TCP/IP, a computer's logical address is called an IP address. An IP address is a number used to identify and locate a computer on a network. Now, how is all of this relevant to client/server applications? It's quite straightforward really; for the client application to talk to a server application, it needs to know where the server application is located, and the only way to establish a connection with a server is through its IP address or host name.

Name Resolution

As we have established before, for the network to identify each computer in its domain, the computers will have network adapters that have unique prefabricated physical addresses, and TCP/IP uses IP addresses to identify a computer on a network. Both of these addressing schemes are numerical and somewhat difficult to remember; they're designed for convenience of the computer (which is very good with numbers), rather than for humans.

For the average person it would be difficult to remember a computer's address is 192.168.0.1 or 131.121.134.146. TCP/IP therefore provides a parallel, user- oriented structure called domain names or DNS names. These names are alphanumeric in nature and easier to remember.

This mapping of IP addresses to domain names is called name resolution. A group of computers called name servers store tables showing how to translate these domain names to and from IP addresses. Email and WWW addresses are expressed as DNS names, for example www.google.com or www.myname.com. Name service systems provide domain name/IP address mappings for DNS registered computers on the network. This ensures that you do not have to enter the IP address of a website to access it.

The most common name resolution method used on the Internet is DNS, but there are others such as Windows Internet Name Services or WINS, which resolves Microsoft Windows NetBios names to IP addresses. How does this relate to the client server application? It provides us with another method of identifying the server location and name. Instead of entering the IP address to access the services provided by a server, we can now enter the domain name, which is easy to remember.