Network analysis

Network analysis

53.1 Purpose

The purpose of this chapter is to overview the key activities in the netawork analysis and design process. Network analysis is a highly specialized discipline in its own right, and a detailed explanation of network analysis tools and techniques is beyond the scope of this book. This chapter is written for systems analysts and information system consultants who must work with network specialists.

53.2 Strengths, weaknesses, and limitations

Not applicable.

53.3 Inputs and related ideas

Network topologies are discussed in Chapter 52. Location connectivity analysis is covered in Chapter 54. Simulation (Chapter 19) and queuing theory (Chapter 79) are two mathematical tools that are often used to support network analysis. The cost estimating process might include a cost/benefit analysis (Chapter 38). Network consultants often develop detailed cost estimates in response to a competitive procurement opportunity (Chapter 41). The need for a network is established during the analysis (Part IV) and high- level design (Part V) stages of the system development life cycle. Key network design parameters are documented in the requirements specifications (Chapter 35).

53.4 Concepts

The purpose of this chapter is to overview the key activities in the network analysis and design process. A detailed explanation of network analysis tools and techniques is beyond the scope of this book.

53.4.1 Requirements identification

The purpose of requirements identification is to ensure that the network designers have clear definitions and a detailed understanding of the essential network requirements and related network design attributes.

Geographical requirements analysis begins with a careful study of the system’s geographical locations (i.e., buildings and areas) and focuses on such issues as topology (Chapter 52) and transmission media (dedicated cable, microwave, satellite, common carrier). Traffic flow pattern analysis is also used to help define the network’s topology (Chapter 52) and connections (Chapter 54) as well as message volumes associated with the various data flows.

Traffic load analysis includes such elements as peak load analysis, message duration analysis, and busy hour analysis. Together, they are used to determine the required number of communication lines, the maximum required capacity for each line, the time slots during which the communication lines are likely to be busy, and several related network performance parameters. The purpose of availability requirements analysis is to determine and document the effect of time differences (time zone shifts) between the different geographical areas covered by the network.

As the name implies, response time analysis is concerned with determining the system’s response time requirements (e.g., interactive, store and forward, real-time, etc.). Reliability requirements analysis provides information that helps the designer develop a back-up plan or create necessary redundancies.

Hardware analysis is used to define the requirements for the personal computers, workstations, terminals, peripherals, communication interfaces, modems, and other hardware that will be attached to the network. Additionally, such software as the operating system and communication protocols must be specified. Future projection analysis focuses on parameters that affect capacity planning, storage requirements, transmission speed, connections with the Internet, and the highway (or Turnpike) effect.

53.4.2 Network design

Based on the network requirements identified in the previous stage, the network is physically designed.

Topology determination focuses on physically laying out the network using such tools as location connectivity diagrams (Chapter 54). The required line speeds are defined based on such criteria as transmitted characters per day, computer time (input, output, and processing) per message, and the required response time. Concentration point determination is concerned with the system’s concentration points (or hubs). Such factors as the number of hubs, the capacities and related requirements for the lines that link the hubs, and the number of clients per server (or per hub) must be weighed against network efficiency. Bottlenecks (or choke points) are places in the network where message flow exceeds capacity, resulting in delays and even lost messages.

Bottleneck analysis (an application of queuing theory, Chapter 79) is a useful, mathematical tool for identifying choke points and for evaluating how various line capacities, transmission speeds, and hardware options (processing speed, storage capacity) affect performance. Queuing theory can also be used to gauge the sensitivity of a network design to such variables (or assumptions) as average message length, message duration, and busy hours. Simulation (Chapter 19) is another useful mathematical tool that can help identify and solve many network design problems. The advantages of using mathematical tools such as queuing theory and simulation include quick feedback, the flexibility to consider numerous variables or assumptions, and low cost.

53.4.3 Cost estimating

Networks are expensive to develop and to operate. Clear, accurate development, and operating cost estimates are essential before the final decision to implement a network is made. The cost estimating process might include a cost/benefit analysis (Chapter 38). Network consultants often develop detailed cost estimates in response to a competitive procurement opportunity (Chapter 41).

53.4.4 Documentation

Once the network analysis is completed, the various design decisions must be carefully documented. Connection diagrams define the topology, connection points, and traffic flows. Key supporting details include the types of transmission media, the desired technology (e.g., T1, ISDN, fiber optics, satellite, microwave), the capacities, speeds, and costs of those media, and a back-up plan. The act of preparing the connection diagrams and supporting documentation is sometimes called logical documentation and preparation.

During the physical documentation and preparation phase, a complete component list is prepared for each node in the network. Key parameters include the brand name, model, speed, and other relevant specifications for such hardware components as the computers, the modems, and related peripherals (printers, scanners, etc.), and the nature and description of all cables and connectors. Sometimes, hierarchy charts (Chapters 48 and 62) are prepared to help document the components that form a subnet work controlled by a hub.

The network specifications incorporate all the information related to the network. In addition to the documentation described above, routers, bridges, and other message switching equipment or devices must be documented in detail. Finally, such details as floor plans, rising cables, distributing cables, central switch boxes, server locations, telephone jack locations, and power outlets must be documented in a wiring diagram.

53.5 Key terms

Availability requirements analysis —

A network analysis process that helps to determine and document the effect of time differences (time zone shifts) between the different geographical areas covered by the network.

Bottleneck (choke point) —

A place in the network where message flow exceeds capacity, resulting in delays and even lost messages.

Bridge —

A computer that links two networks with similar protocols.

Client —

A computer (more generally, a node) that requests a service from a server.

Client/server —

A network in which client computers request services from a central server computer.

Concentration point determination —

A network analysis process that is concerned with the system’s concentration points (or hubs).

Connection diagram —

A diagram that shows the topology, connection points, traffic flows, and patterns of a network.

Data communication —

The act of transmitting data from one component to another.

Distributing cable —

Generally, a cable that links the computers or nodes on a single floor.

Future projection analysis —

A network analysis process that focuses on parameters that affect capacity planning, storage requirements, transmission speed, connections with the Internet, and so on.

Gateway —

A computer that links two or more networks with different protocols.

Geographical requirements analysis —

A preliminary network analysis process that begins with a careful study of the system’s geographical locations and focuses on such issues as topology and transmission media.

Hardware analysis —

A network analysis process that helps to define the requirements for the personal computers, workstations, terminals, peripherals, communication interfaces, modems, and other hardware that will be attached to the network, and such software as the operating system and communication protocols.

Highway effect (turnpike effect) —

The tendency of users to quickly adopt new technology as soon as it proves its usefulness; because of the highway effect, the demands placed on a system often exceed projections. This term was initially coined in the 1950s when the traffic load on the Pennsylvania Turnpike exceeded the designers’ long-term, worst-case projections soon after the road opened.

Host —

A computer in a wide area network.

Hub —

A central controlling device, point, or node in a network.

Local area network (LAN) —

A network in which the nodes are located in close geographic proximity and are generally linked by direct lines.

Message switching —

The process of routing a message from its source to its destination; note that sometimes messages are decomposed into packets that reach their destination via different transmission paths.

Network —

Two or more computers linked by a communication line.

Network topology —

A map of a network; a physical arrangement of the nodes and connections in a network.

Node—A connection point (computer, workstation, peripheral, concentrator, etc.) in a network.

Protocol —

A set of rules that governs data communication.

Reliability requirements analysis —

A network analysis process that helps the designer develop a back-up plan or create necessary redundancies.

Response time analysis —

A network analysis process that helps to determine the system’s response time requirements (e.g., interactive, store and forward, real-time, etc.).

Rising cable —

Generally, a cable that runs between two floors in a building.

Router —

An intelligent device that provides network connections and performs such services as protocol conversion and message routing.

Server —

A computer that holds centralized resources and provides them to clients on request.

Token passing —

A network management technique in which an electronic token is passed continuously from node to node around the network and a given node can transmit a message only when it holds the token.

Topology —

A map of a network; a physical arrangement of the nodes and connections in a network.

Topology determination —

A network analysis process that focuses on physically laying out the network using such tools as location connectivity diagrams.

Traffic flow pattern analysis —

A network analysis process that helps to define the network’s topology and connections as well as the message volumes associated with the various data flows.

Traffic load analysis