Dennis Hall Closet

    This is a diagram of the Dennis Hall Electrical Closet. In this closet are the active components, patches, cables, and connections that service Dennis Hall. Please note that this diagram of the closet looks tidy and neat, but in reality this closet is probably the messiest closets on campus. Nevertheless, it serves as a good representation of most electrical closets. It also serves as a good example of how a electical closet should not be designed.
    The multi-mode fiber comes in through a conduit and into what's labeled "Jumper box." Inside the box, the fibers are terminated onto the ends of plugs. The multi-mode fiber jumpers then connect the other end of the plugs to the hubs. Notice that there are two sets of hubs. One set services the faculty subnet and the other set services the public subnet. I talk a bit about subnets later.
    The hubs used in this closed are probably the dumbest components used in the Earlham College network. All they do is take the signals that come in from the fiber or the ports they service and repeat the signals. So, if one of the faculty sends a request to WebDB, then the hub receives the signal and propagates the same signal through the fiber. What could be simpler. Hubs are considered to be active components because they need to be plugged into an electrical socket. In fact, any component that needs to be plugged into a power source is an active component.
    The hubs connect to the ports to which the indivisual terminals are connected via the punch board (or whatever you like to call it). A CAT twisted-pair jumper pluges into one of the port on the hub and the other end connects to a RJ45 plug on the punch board. From the punch board, a CAT 5 twisted pair cable runs to a port, where the cable is punched into another RJ45 connector. Another twisted-pair jumper runs from the port to an ethernet card which is installed in the terminal.
    Almost all electrial closets on campus are very similar to the Dennis Electrical Closet. How many sets of hubs are used depends on the number of subnets within a building are used. For example, Bundy is a residence hall, therefore there are not any administrative offices, faculty offices, etc. Thus, Bundy only has one set of hubs that service the student subnet. Carpenter has administrative offices, and faculty offices. Thus, there are hubs for the administrative and faculty subnets. There may even be public subnets for the classrooms.     A subnet is a logical division of the network. Subnets are used to control traffic so that packets to not have to go where they don't have to. It also adds to the security of network traffic. For example, packets that goes between faculty computers usually does not contain information that would be of use to students. The information probably should not be seen by students. So, it makes sense to keep faculty traffic separate from student traffic.   If a faculty member needs to send email to a student, then the router will redirect the packet from the faculty to student subnet. Intra-subnet travelling is easily done with a router.