Basic Network Configuration

The International Organization for Standards (ISO) established the protocol layers of the Open Systems Interconnection (OSI) in 1978. The importance of this model is that it influences the client/server architecture and the network operating systems of the networked environment. The OSI model has seven distinct layers and these layers can be grouped into two to show what each layer does on the network. The OSI model basically breaks up the job of moving the data packets from one point to another into seven different tasks.

OSI Model

	Protocol Layers	Activity
End User Group	Application (7)	User programs: initiate and process messages.
	Presentation (6)	System utilities: standardize data appearance, text compression.
	Session (5)	Operating system: establish user level session and manage sessions between applications; message sequencing, recovery.
	Transport (4)	Network manager: flow control, end to end error detection and correction, priority service.
Network Group	Network (3)	Network manager: manage connection, routing, message blocking into packets.
	Data link (2)	Hardware: reliable data delivery over physical medium; transmission error recovery, separating packets into frames.
	Physical (1)	Hardware: actual communication across physical medium; individual bit communication.

(Pfleeger, 1997, 382)

Hardware would encompass the first three layers; Physical, Data link and part of Network. Operating system would continue from part of the Network layer and through the Transport, Session and part of the Presentation layers. System utilities deal specifically with the Presentation layer and Application programs with the Application layer of packet. The network protocols ensure that the data travels through the network quickly and accurately while the end-user protocols ensure that the distribution, exchange, formatting, and transmission of the data is performed properly (Eckel & Steen, 1996). The OSI model is less a specification and more a framework for network designers to build their architecture. Thus, the overlapping of the layers is possible but the basic structure and logic for how the architectures are put together is the same (Campbell, 1996).

The client/sever computing approach to network architecture uses many of the layers from the OSI model in its distribution of functions.

Application. Top most layer of the architecture. It performs the management of application processes, distribution, of data, interprocess communication, and decomposition of application functions into distributable processes. Application layer functionality is supported by lower layers.

Distributed Operating System Layer. This layer provides the system-wide distributed services required by the application layer. It supports global naming, directory, addressing, sharing of local resources, protection and synchronization, intercommunication and recovery.

Local Management and Kernel Layer.This layer supports the distributed operating system in the individual nodes. It supports local interprocess communication memory and I/O access, protection, and multitasking.

Communication System Layer. This layer supports communications required by the application, distributed operating system, and local management layers. (Eckel, 1996, 23)

With this layered architecture there are several benefits to be had. Eckel (1996) describes the benefits as layer independence, flexibility, simplified implementation and maintenance, and standardization. What is a client/server architecture in layman's terms then?

A client is a computer process that requests services from network resources. A server is a computer process that provides services to authorized requesters. Client/server refers to a computing architecture, not a given technology (Bennett, 1996, 76).

For large organizations like libraries, client server architecture has several advantages to other network architectures. Flexibility is afforded client/server networks because of the options available to them for upgrading and more software is written specifically for them. In addition to these there is also greater industry information regarding the installation and configuration of such networks, and there is the administrative control and the expandability of the network to handle huge numbers of nodes.

Intranets require a special protocol in order to access servers that may not be the same in terms of NOS. The foundation of the Internet and Intranets is the Transmission Control Protocol/Internet Protocol (TCP/IP) family of network protocols. "A protocol is a set of rules that define two things - the format of the packets and the semantics of their use" (Eckel,1996, 61). Packets are the electronic units of information that are transmitted across the network and is the concern of the OSI Model. The packet contains the sending node's address, the receiving node's address and the data being sent between the two nodes. TCP/IP is a non-proprietary protocol suite that allows transparent communication to occur between separated computers and exchange information. It was established as a standard by the Internet Engineering Task Force (IETF) to allow communication to take place over the Internet with no licensing costs (Bennett, 1996). As a result of the non-proprietary nature of TCP/IP many Internet/Intranet applications are based on this protocol suite and have made the phenomenon of both so prolific (Guengerich, 1997).

HyperText Transfer Protocol is a "communication model that could be used by clients and servers to exchange hypertext documents over a network" (Bernard, 1996, 54), and is essential to any Intranet because it allows the client and server easy communication and transmission of hypertext files. The transmission of files is done by using a standard request format called a Universe Resource Locator (URL). The URL is a path name to the server which contains the directory web pages that an individual is looking for. When this is applied to an Intranet environment it allows for easy access to files through what would now be referred to as a Web Server and the clients, or browsers, would issue the request of the URL. Upon receipt of the request the client (browser) would interpret the and display the file for the user (Bennett, 1996). The W3C's definition of HTTP is as follows:

HTTP is an application level protocol with the lightness and speed necessary for distributed, collaborative, hypermedia information systems. A feature of HTTP is the typing and negotiation of data representation, allowing systems to be built independently of the data being transferred
(Bennett, 1996, 78)

You could almost replace the term HTTP with Intranets in this definition to help describe what Intranets do.

Contents

Table of Contents

Part 1: Intranets Philosophy Structure Part 2: Intranet Installation Basic Network Configuration Server Client Publishing Applications Security Conclusion References