Information Technology Reference
In-Depth Information
SIMULATION GAMING
Simulation gaming is predicated on the notion that students learn better from experience
than from lectures and reading. It also introduces a sense of play into learning, which makes
it fun, as it should be. Students are actively involved rather than passive, and the connections
between pieces are often easier to see when experienced (Corbeil, 1989; Greenblat, 1988). After
many years of experience using various games in class, the author has come to believe
strongly in their value as pedagogical vehicles (see, for example, Butterfield & Pendegraft,
1996; Pendegraft & Watson, 1990). Informal student response to these games has been
enthusiastic. Here, a simulation game was developed to let the students experience the
operation of TCP/IP. The purpose of the simulation was to help the students understand the
function of each layer and, in particular, to understand how the layers interact with each other.
TCP/IP
There are many excellent discussions of TCP/IP (for example, Hunt, 1998). This section
is not intended to give a detailed or technical description of TCP/IP, but rather to clarify some
of the simplifications imbedded in the simulation. In order to keep the simulation manageable,
many of the specific functions are excluded from the simulation or are included only
notionally: that is, they are dealt with via conversation between the students and the
instructor.
TCP/IP is a layered packet switching protocol. Messages are broken into pieces
(packets), each of which is sent to the destination independently of the other pieces. Each
layer performs a set of functions, which together, result in a reliable communications
connection. Each layer adds to the packets its own header containing information needed
by the same layer at the receiving end. While each layer may take the packet from the higher
layer and break it into smaller pieces, this is only done by the TCP layer in the simulation. Error
correction is handled notionally. This simplifies the simulation and allows for focus on more
important issues. A brief summary of each TCP layer follows.
Application Layer
The application layer includes processes like email, Web services, and database
management systems. A port number identifies each application running on top of TCP/IP.
Several well-known port numbers are used in the example.
The Domain Name Service (DNS) is an application layer service that converts domain
names like www.whitehouse.gov or ebay.com into their IP addresses. DNS services are
outside the scope of this exercise and are notional in the simulation. DNS information is
assumed available to IP in the simulation.
TCP Layer
The transport control protocol (TCP) creates and maintains connections between
machines. It breaks traffic into pieces called segments, calculates and appends a CRC (cyclical
redundancy checksum) to each, and sends them to the next layer, IP. Each packet is addressed
to the correct port (application) on the receiving machine. TCP on the receiving machine
checks the CRC of each segment and acknowledges those correctly received. Those
unacknowledged in time are resent. Error detection and retransmission are handled notionally
