- OSI Reference Model -
Network Reference Models
As computer network communication grew more prevalent, the need for a
consistent standard for vendor hardware and software became apparent.
Thus, the first development of a network reference model began in the
1970’s, spearheaded by an international standards organization.
A network reference model serves as a blueprint, dictating how network
communication should occur. Programmers and engineers design products
that adhere to these models, allowing products from multiple manufacturers
to interoperate.
Network models are organized into several layers, with each layer assigned
a specific networking function. These functions are controlled by protocols,
which govern end-to-end communication between devices.
Without the framework that network models provide, all network hardware
and software would have been proprietary. Organizations would have been
locked into a single vendor’s equipment, and global networks like the
Internet would have been impractical or even impossible.
The two most widely recognized network reference models are:
• The Open Systems Interconnection (OSI) model
• The Department of Defense (DoD) model
The OSI model was the first true network model, and consisted of seven
layers. However, the OSI model has become deprecated over time, replaced
with more practical models like the TCP/IP (or DoD) reference model.
Network models are not physical entities. For example, there is no OSI
device. Devices and protocols operate at a specific layer of a model,
depending on the function. Not every protocol fits perfectly within a specific
layer, and some protocols spread across several layers.
The Open Systems Interconnection (OSI) model was developed in the
1970’s and formalized in 1983 by the International Organization for
Standardization (ISO). It was the first networking model, and provided the
framework governing how information is sent across a network.
The OSI Model (ISO standard 7498) consists of seven layers, each
corresponding to a particular network function:
7 Application
6 Presentation
5 Session
4 Transport
3 Network
2 Data-link
1 Physical
Various mnemonics have been devised to help people remember the order of
the OSI model’s layers:
7 Application All Away
6 Presentation People Pizza
5 Session Seem Sausage
4 Transport To Throw
3 Network Need Not
2 Data-link Data Do
1 Physical
The ISO further developed an entire protocol suite based on the OSI model;
however, this OSI protocol suite was never widely implemented. More
common protocol suites can be difficult to fit within the OSI model’s layers,
and thus the model has been mostly deprecated.
A more practical model was developed by the Department of Defense
(DoD), and became the basis for the TCP/IP protocol suite (and
subsequently, the Internet). The DoD model is explained in detail later in
this guide.
The OSI model is still used predominantly for educational purposes, as
many protocols and devices are described by what layer they operate at.
The Upper Layers
The top three layers of the OSI model are often referred to as the upper
layers. Thus, protocols that operate at these layers are usually called upperlayer
protocols, and are generally implemented in software.
The function of the upper layers of the OSI model can be difficult to
visualize. The upper layer protocols do not fit perfectly within each layer;
and several protocols function at multiple layers.
The Application layer
The Application layer (Layer 7) provides the actual interface between the
user application and the network. The user directly interacts with this layer.
Examples of application layer protocols include:
• FTP (via an FTP client)
• HTTP (via a web-browser)
• SMTP (via an email client)
• Telnet
The Presentation layer
The Presentation layer (Layer 6) controls the formatting of user data,
whether it is text, video, sound, or an image. The presentation layer ensures
that data from the sending device can be understood by the receiving device.
Additionally, the presentation layer is concerned with the encryption and
compression of data.
Examples of presentation layer formats include:
• Text (RTF, ASCII, EBCDIC)
• Music (MIDI, MP3, WAV)
• Images (GIF, JPG, TIF, PICT)
• Movies (MPEG, AVI, MOV)
The Session layer
The Session layer (Layer 5) establishes, maintains, and ultimately
terminates connections between devices. Sessions can be full-duplex (send
and receive simultaneously), or half-duplex (send or receive, but not
simultaneously).
The four layers below the upper layers are often referred to as the lower
layers, and demonstrate the true benefit of learning the OSI model.
The Transport Layer
The Transport layer (Layer 4) is concerned with the reliable transfer of
data, end-to-end. This layer ensures (or in some cases, does not ensure) that
data arrives at its destination without corruption or data loss.
There are two types of transport layer communication:
• Connection-oriented - parameters must be agreed upon by both
parties before a connection is established.
• Connectionless – no parameters are established before data is sent.
Parameters that are negotiated by connection-oriented protocols include:
• Flow Control (Windowing) – dictating how much data can be sent
between acknowledgements
• Congestion Control
• Error-Checking
The transport layer does not actually send data. Instead, it segments data
into smaller pieces for transport. Each segment is assigned a sequence
number, so that the receiving device can reassemble the data on arrival.
Examples of transport layer protocols include Transmission Control
Protocol (TCP) and User Datagram Protocol (UDP). Both protocols are
covered extensively in another guide.
Sequenced Packet Exchange (SPX) is the transport layer protocol in the
IPX protocol suite.
The Network Layer
The Network layer (Layer 3) has two key responsibilities. First, this layer
controls the logical addressing of devices. Logical addresses are organized
as a hierarchy, and are not hard-coded on devices. Second, the network layer
determines the best path to a particular destination network, and routes the
data appropriately.
Examples of network layer protocols include Internet Protocol (IP) and
Internetwork Packet Exchange (IPX). IP version 4 (IPv4) and IP version 6
(IPv6) are covered in nauseating detail in separate guides.
The Data-Link Layer
The Data-Link layer (Layer 2) actually consists of two sub-layers:
• Logical Link Control (LLC) sub-layer
• Media Access Control (MAC) sub-layer
The LLC sub-layer serves as the intermediary between the physical link and
all higher layer protocols. It ensures that protocols like IP can function
regardless of what type of physical link is being used.
Additionally, the LLC sub-layer can use flow-control and error-checking,
either in conjunction with a transport layer protocol (such as TCP), or
instead of a transport layer protocol (such as UDP).
The MAC sub-layer controls access to the physical medium, serving as
mediator if multiple devices are competing for the same physical link.
Specific technologies have various methods of accomplishing this (for
example: Ethernet uses CSMA/CD, Token Ring utilizes a token).
The data-link layer packages the higher-layer data into frames, so that the
data can be put onto the physical wire. This packaging process is referred to
as framing or encapsulation. The encapsulation type used is dependent on
the underlying data-link/physical technology (such as Ethernet, Token Ring,
FDDI, Frame-Relay, etc.)
Included in this frame is a source and destination hardware (or physical)
address. Hardware addresses usually contain no hierarchy, and are often
hard-coded on a device. Each device must have a unique hardware address
on the network.
The Physical Layer
The Physical layer (Layer 1) controls the transferring of bits onto the
physical wire. Devices such as network cards, hubs, and cabling are all
considered physical layer equipment.
Physical-layer devices are covered extensively in other guides.
Explanation of Encapsulation
As data is passed from the user application down the virtual layers of the
OSI model, each of the lower layers adds a header (and sometimes a
trailer) containing protocol information specific to that layer. These headers
are called Protocol Data Units (PDUs), and the process of adding these
headers is called encapsulation.
For example, the Transport layer adds a header containing flow control and
sequencing information (when using TCP). The Network layer header adds
logical addressing information, and the Data-Link header contains physical
addressing and other hardware specific information.
The PDU of each layer is identified with a different term:
Layer PDU Name
Application -
Presentation -
Session -
Transport Segments
Network Packets
Data-Link Frames
Physical Bits
Each layer communicates with the corresponding layer on the receiving
device. For example, on the sending device, hardware addressing is placed
in a Data-Link layer header. On the receiving device, that Data-Link layer
header is processed and stripped away before it is sent up to the Network
and other higher layers.
Specific devices are often identified by the OSI layer the device operates at;
or, more specifically, what header or PDU the device processes. For
example, switches are usually identified as Layer-2 devices, as switches
process hardware (usually MAC) address information stored in the Data-
Link header of a frame.
Similarly, routers are identified as Layer-3 devices, as routers look for
logical (usually IP) addressing information in the Network header of a
packet.
OSI Reference Model Example
The following illustrates the OSI model in more practical terms, using a web
browser as an example:
• At the Application layer, a web browser serves as the user interface for
accessing websites. Specifically, HTTP interfaces between the web
browser and the web server.
• The format of the data being accessed is a Presentation layer function.
Common data formats on the Internet include HTML, XML, PHP, GIF,
JPG, etc. Additionally, any encryption or compression mechanisms used
on a webpage are a function of this layer.
• The Session layer establishes the connection between the requesting
computer and the web server. It determines whether the communication
is half-duplex or full-duplex.
• The TCP protocol ensures the reliable delivery of data from the web
server to the client. These are functions of the Transport layer.
• The logical (in this case, IP) addresses configured on the client and web
server are a Network Layer function. Additionally, the routers that
determine the best path from the client to the web server operate at this
layer.
• IP addresses are translated to hardware addresses at the Data-Link
layer.
• The actual cabling, network cards, hubs, and other devices that provide
the physical connection between the client and the web server operate at
the Physical layer.
IP and the DoD Model
The Internet Protocol (IP) was developed by the Department of Defense
(DoD) during the late 1970’s. It was included in a group of protocols that
became known as the TCP/IP protocol suite.
The DoD developed their own networking model to organize and define the
TCP/IP protocol suite. This became known as the DoD Model, and consists
of four layers:
OSI Model DoD Model
7 Application
6 Presentation
5 Session
4 Application
4 Transport 3 Host-to-Host
3 Network 2 Internet
2 Data-link
1 Physical
1 Network Access
The DoD model’s streamlined approach proved more practical, as several
protocols spread across multiple layers of the OSI Model.
The following chart diagrams where protocols fit in the DoD model:
Layer Example Protocols
Application FTP, HTTP, SMTP
Host-to-Host TCP, UDP
Internet IP
Network Access Ethernet