The OSI Model

Uploaded by itfreetraining on 25.07.2011

In this section I will look at the OSI model. To understand networking it payees to have
an understanding of the OSI model as parts of it are used in modern networking.
The OSI model, or Open System interconnection model, is a network model designed back in
the late 1970’s. The model itself did not get adopted, however references are still
made to it in modern networking especially to the names and the numbers in the OSI model.
You will find that most network technologies do not match exactly to the OSI model and
do not provide an exact fit, however it is useful when discussing networking to reference
it against the osi model. The OSI model has 7 layers. Each layer communicates
with the layer above and below it only. By doing this you can change a layer with effecting
other layers as long as the new layer is compatible with the adjacent layers.
To understand the OSI model better and how it works, consider a manufacturing plant.
Image the first step of the manufacturing plan is to make a part. The next step put
a protective coating on the part. After this the part is put in a box. The box than has
a shipping label placed on it. The important thing to consider is that each process of
the manufacturing process works on the previous. For example you could not apply the coating
on the part if you had already place the part in the box. The OSI models works the same,
each process or layer of the OSI model adds to the previous.
The OSI model consists of seven layers. Each layer handles a different part of network
communication. The first is the application layer. This is where the application programming
interface or API’s are found. Applications call functions in theAPI to sent network traffic.
Next you have the presentation layer. This layer coverts the data inputted from
the application into a form suitable for transmission over the network. The session layer controls
communication between two applications. For example the session layer may decide on the
port used for communication. The transport layer ensure the data arrives in the same
order it was sent and is not duplicated. The network layer determines the path used
to transfer data across the network. The Data link layer provides error checking and creates
the packet to be sent over the network. Lastly the physical layer sends the data over the
network media to the other party. To remember the seven layers of the OSI model
you may want to consider using a mnemonics. Two popular ones are all people seem to need
data processing and please do not tell sales people anything. Take the first letter of
each word to helps you remember each layer name. For A is for Application, P is for presentation,
S is for session and so on. The second mnemonics works from the bottom up, for example P is
for physical, D is for Data Link, N is for network and so on.
To better understand how the OSI model works, consider this. If a computer want to sent
a message to anther computer on the network it communicates with the application layer.
The application layer communicates with the presentation layer. The communication goes
all the way down to the physical layer. As the data go down the layers more information
is added like packet headers and checksums. The data is than communicate across the network
over the physical layer. The physical layer will check the checksum
and pass the data up to the data link layer. As the data goes up the layers more data is
removed like packet headers. When the data gets to the top, all the headers and check
sums are removed and the data is presented to the application in a way that is can understand.
As stated before the OSI model in it’s entirety was not adopted however the layers names and
layer numbers are still used today. TCP IP is divided into 4 layers and can be roughly
mapped to the OSI model. To better understand why TCP is divided into 4 layers it so a good
idea to look at what happens at each layer. At the application layer you have protocols
like http, smtp, d n s and RIP. At the next layer you have t c p and U D P. Using T C
P you have reliable transmission, that is packets are guarantee to be delivered and
delivered in order. U D P uses unreliable transmission and does not guarantee delivery
or order of delivery. At the internet layer you have the choice
of ip version 4 or ip version 6. The final layer provides a link to your network media.
As you can see using layers allows you to add and remove protocols and services very
easily. For example you could add a new network card, for example a token ring network card
to your computer and as long as the new network card can communicate correctly to the layer
above nothing else needs to be changed. The new network card will work with any protocols
at the application layer and any transports at the transport layer without any changes.
As you can see, using layers makes adding and subject hardware, protocols and services
easy. To put the OSI model in prospective, you will
see in a lot of literature that refers to a device as a layer 2 device or layer 3 device.
Switches work on layer 2 of the OSI model. If you consider this small network. The switch
is able to decide based on the destination packet which device to sent it to. For example,
packets coming from the computer to the printer would be directed straight to the printer.
Packets going to the server from the computer will not be seen by the printer. The switches
looks at the destination address or mac address to determine where to sent the packet.
If you add anther network separated by a wide area network, the layer 2 switch will not
be able to router traffic to this network. To router traffic between different networks
you need a layer 3 device for example a router. The router will be able to router traffic
between the two network as routing is a function on layer 3 of the OSI model. You may see devices
on the market called a layer 3 switches. Don’t get confused by this, remember layer 3 devices
support routing. Simply put, a layer 3 switch supports routing and switching. Layer 3 switches
performing routing in hardware rather than software so tend to be a faster than routers
at routing traffic between networks. Since Layer 3 switches logic is build into hardware
they lack some features of routers. Keep this in mind if you decide to deploy one on your
network. Just remember this, layer 2 devices perform
local switching in a local network. Layer 3 devices supports routing between networks.
Devices like firewalls can work on much high levels of the OSI model. For example a good
firewall can work all the way up to the application level and detect threats in protocol’s like
http. If you see a device that says it works on
a particular level of the OSI model, think what that level does and that will tell you
what the device does. For the exam, make sure understand the basics of how the OSI model