Networking I: Open System Interconnect (OSI) Reference Model

By Ruffin Scott, ACI Technical Support

Technical Note 99-28

Technical Notes for Technical Notes for 99-07-July 1999

4th Dimension developers are often called upon to help manage or troubleshoot a client's network, or internetwork. It is important to understand the basic concepts of networking so that installation and operation of 4th Dimension products goes smoothly.

This is the first in a series of technical notes that will address networking. This technical note outlines the fundamental concepts of the OSI Reference Model, and defines some common networking terms.

A network is two or more nodes connected so that they can share information. The most common type of network is the local area network(LAN), in which all the nodes reside in the same geographic location. A typical LAN is a small office that has two or more computers that are connected and share a common printer.

LANs originated with the emergence of the personal computer. They filled the need for PC users in close geographical proximity to share resources and information. Shared access to file servers, application servers, printers, and the ability to exchange information directly, are all economic benefits of the LAN.

One of the best ways to understand networking is to study the process by which computers and other networking devices transfer data. Developed by the Organization for Standardization (ISO) in 1984, the OSI model conceptually describes the foundation of how data is transferred from an application on one computer to an application on another. The OSI model comprises seven different parts called layers. Each layer handles a specific function in transferring data through a network. The layers are generally separated into two groups -- layers one through four are considered the "Lower" layers, while layers five through seven are the "Upper" layers. The lower layers transport data and the upper layers manage application specific tasks. The seven layers are as follows:

The highest layer in the OSI model, the Application layer interfaces directly with software applications that require communication with other computers or devices. At this level, the software will establish availability of network resources and other devices with which it can communicate, as well as handle communication synchronization. TCP/IP protocols such as File Transfer Protocol (FTP) and Simple Mail Transfer Protocol (SMTP) are implemented at the Application layer.

The Presentation layer transforms data from the Application layer and formats it so that the Application layer of the receiving device can interpret the data. Encryption, image and video formatting, character translation, and compression are all common functions at the Presentation layer.

The Session layer controls communication between the Presentation layers of two or more networked computers or devices.

The Transport layer is the highest layer of the "Lower" layer group. One of its major functions is to handle data transmission flow control. By managing flow control, this layer prevents more data from being sent or received by a computer or device than it can handle at any given instant. Transmission Control Protocol (TCP), which establishes data transmission reliability, is implemented at this layer.

Routing protocols such as Routing Information Protocol (RIP), Border Gateway Protocol (BGP), and Open Shortest Path First (OSPF) are implemented at the Network layer. These routing protocols manage data routing through an internetwork. Routers are most commonly utilized within the Network layer

The Data Link layer is split into two sublayers. The upper sublayer is the Logical Link Control (LLC), which frames packets of data from the upper layers for transmission. The lower sublayer is the Media Access Control (MAC), which handles the addressing of the frames before they are sent to the Physical layer.

The Physical layer is the physical medium, i.e. cabling/hardware, by which the data is transmitted from one network node to another. Transmission distances, data rates, and network connectors are all defined by the Physical layer.

The general process of how the OSI model defines data transmission from one node to another is fairly simple. Each layer interacts with the layer above and below on the same node, and with the same layer on another node. The following diagram shows this interaction.

To describe the networking process, we use a simplified example of computer A sending an email to computer B. The person at computer A uses an email application to write the email at the Application layer. When an email is sent, it is passed to the Presentation layer where it is formatted so computer B's Presentation layer can understand it. From there, the Presentation layer passes it to the Session layer. The Session layer on computer A manages the transmission session with computer B. The data is then passed to the Transport layer, which handles flow control between the two computers.

The Transport layer appends a header to the data and passes it to the Network layer. After the Transport layer has attached its header, the data is now called a Datagram. The Network layer then appends its own header and manages the routing of the data. At this point, the data is called a packet. The packet is passed to the Data Link layer, which appends its header and frames the packet for transmission. The frame is sent to the Physical layer where it is converted to electrical voltages and sent via cable to computer B.

Computer B receives the packet at the Physical layer and passes it up to the Data Link layer. The Data Link layer strips out its header, and passes the data to the Network layer. The Network layer strips out its header and passes the data to the Transport layer. The Transport layer strips out its header and sends the data to the Session layer. The Session layer passes the data to the Presentation layer. The Presentation layer removes the formatting from the data and passes it to the Application layer. The Application layer displays the email to the user at computer B.

The following diagram shows this process.

Internetworks are multiple networks that are connected in such a way that they act as one large network, connecting multiple office or department networks. Internetworks are connected by networking hardware such as routers, switches, and bridges. An example of an internetwork is two office branches connected together by a T1 line.

Internetworking is a solution born of three networking problems: isolated LANs, duplication of resources, and the lack of a centralized network management system. With connected LANs, companies no longer had to duplicate programs or resources on each network. This in turn gave way to managing the network from one central location instead of trying to manage each separate LAN.

A WAN is a network where one or more nodes of the network are in different geographic locations. An example of a WAN is a sales staff in various regions dialing into the main office.

Router: A networking device that operates at Layer 3 of the OSI model. Routers are used to mix media types and are most often used to make WAN connections. Routers can handle multiple protocols (IP, IPX, AT, etc.) and use routing protocols to determine the best path through a network. Routers are sometimes referred to as gateways.

Switch: A networking device that operates at layer 2 of the OSI model. Switches are used to provide connectivity to users on a LAN and to improve network performance over shared environments. Switches can be small devices or can scale to have hundreds of connections.

Hub: Typically used in Ethernet environments. Hubs are used to enable multiple devices to share one 10MB or 100MB connection. Hubs do not improve network performance -- all users are still on one collision domain.

Firewall: A device (router or other hardware/software combination) used to separate a public and private domain. Typically, firewalls are used to protect network assets from invasion by Internet users.

Collision domain: Refers to a "shared" Ethernet segment. Hubs enable you to add more users to a single collision domain. Switches enable you to segment collision domains.

Ethernet: Popular LAN technology based on the IEEE 802.3 standard. Ethernet can run in 10Base, 100Base and 1000Base speeds. It uses CSMA/CD (Carrier Sense Multiple Access Collision Detection) for its media detection method. Unlike token ring, it is possible for two users to transmit at the same time -- resulting in a collision.

Token Ring: A token-passing LAN technology based on IEEE 802.5. Token Ring operates at 4 or 16Mbps. Connectivity is provided by an MSAU (Multi-Station Access Unit).

FDDI: Fiber Distributed Data Interface. FDDI is a token-passing LAN media that runs at a speed of 100Mbps for distances up to 2 kilometers.

CDDI: Copper Distributed Data Interface. CDDI is a variation of FDDI that is run over copper cable. CDDI runs at a speed of 100Mbps for distances up to 100 meters.

ATM: Asynchronous Transfer Mode. ATM is standard for cell relay in which multiple service types (such as voice, video, or data) are carried in fixed-length cells. Fixed length cells allow cell processing to occur in hardware, thereby reducing transit delays -- ATM cells are 53 bytes long. ATM is designed to run on high speed interfaces such as DS-3, OC-3, etc.

Frame Relay: Popular WAN technology that handles multiple circuits through one physical connection. Allows one WAN connection on a router to connect with multiple remote sites. Frame relay is a packet switched network that determines the path through the frame relay cloud (internetwork) based on PVC information.

Leased Line: A dedicated WAN connection. This is a private connection with guaranteed throughput between sites.

ISDN: Integrated Services Digital Network. A WAN connection that can carry voice, data and video. ISDN is offered as a BRI or a PRI. BRI (Basic Rate Interface) has two 64k B channels and a 16k D channel. PRI (Primary Rate Interface) has twenty-three 64k B channels and one 24k B channel. B channels are used to carry voice, data, video traffic; D channels are used to establish and tear down connections.

DSL: Digital Subscriber Line. High speed public network media that provides high bandwidth over standard phone lines. DSL is limited by distance. It requires a DSL "modem" at the central office and at the customer premises. DSL does not use the entire bandwidth of a wire, leaving room for a voice connection.

T1/E1: T1 lines transmit at 1.544Mbps through the public telephone network, and are typically used in North America. E1s transmit at 2.48 Mbps through the public telephone network, and are typically used in Europe and South America.

T3/E3: T3s transmit at 44.736 Mbps, and are typically used in North America. E3s transmit at 34.368 Mbps, and are used in Europe and South America.

This technical note presents the basic concepts of the OSI Reference Model to offer a foundation for understanding more difficult networking concepts. It also addresses common networking terms and devices. The other technical notes in this series will explain these terms and devices in further detail.