Traditionally computer network designers developed their systems with the assumption that they will operate on a specific type of cable using a specific type of connector. Thus, each manufacturer has its own cable and connector “standard,” but that standard was for their own system only. For example, here are some “standard” cable/connector systems in use:
|IBM S/3x and AS/400||100 Ohm Twinax & Twinax Connectors|
|IBM 3270||93 Ohm Coax & BNC Connectors|
|IBM Token Ring||150 Ohm Shielded Twisted Pair & IBM Data Connector|
|Hewlett Packard 3000||RS-232 Cable & DB Connectors|
|Ethernet||50 Ohm Coaxial Cable & BNC or N Connectors|
|Wang||Dual 75 Ohm Coax & BNC-TNC Connectors|
So, when migrating from one type of computer system or network to another, the entire cabling system must be abandoned, and an entirely new cabling system must be installed, which may cost more than the hardware itself.
Another problem is the cost of making moves, changes or additions after the original installation. For example, IBM’s System/3X and AS/400 uses “daisy chain” or “bus” topology, which makes changes difficult because cables either have to be moved, extended, or added. With large networks, this wastes time and money because:
- It is difficult or impossible to migrate from one computer system to another without replacing the entire cabling system, and
- To make moves, changes or additions, the cabling system has to be changed. So there is no real structure since it is constantly changing as user requirements change, making it “non-structured.”
By installing “structured” wiring systems, where all cables are run to one central location and pre-wiring all possible locations, needed changes can be made quickly by simply moving patch cables around in a centralized wiring closet. Simply by changing the cables and devices at the equipment and workstation, creating a “structured” cabling system that can be adapted to a variety of systems and interfaces.