Understanding Campus Internetworks

The history of networking is a history of ebbs and flows. From the initial networks, which were designed to provide access to simple central, shared resources on the mainframe computer, we moved to the distributed architecture of networks in the 1990s. This has been followed by a move toward server farms, which in many ways appear to be a return to the old centralized networking from the past.

Mainframes were not always discarded; some still carry out huge batch processing tasks in banks and insurance companies, but many just became storage areas for data and databases. The NetWare or NT server took over as a file/print server and soon started running most other programs and applications as well. Groups of servers running sympathetic applications were clustered together in domains, or other administrative groups, and new directory services emerged to allow easy discovery of domain services. Networks were developed to find the simplest, cheapest, and most reliable mechanisms to establish and maintain connectivity with the resources.

Over the last 20 years, we have witnessed the birth of the LAN and the growth of WANs (Wide Area Networks) and the Internet. More than anything else, the Internet is changing our lives daily, with ever-increasing numbers of online transactions taking place, education and entertainment services becoming available, and people just plain having fun communicating with each other in exciting new ways.

So how will networks evolve in the twenty-first century? Are we still going to see file and print servers at all branch locations, or will servers migrate to common locations? Are all workstations going to connect to the Internet with ISPs to separate the data, voice, and other multimedia applications? I wish I had a crystal ball.