The OSI World

It won’t take you too long being in a technology world to realize there is a concept out there called the OSI Model that people keep referencing again and again and again. This mythical concept that every technology personnel use to categorize layers and functionality is a foundational reality of connectivity. Thus what I would like to do in this article is to walk you through what the OSI Model is all about conceptually, then apply it to real life.

If you haven’t seen it before, let me be the first to welcome you to this mythical concept named the OSI model, all seven layers of its glory. Your number one task is to memorize it. My favorite acronym to remember the layers is the phrase “Please Do Not Throw Sausage Pizza Away”, which is from the bottom up, or “All People Seems To Need Data Processing”, moving from top to bottom, or may be you can come up with a better one.

While you are busy memorizing it, let me describe the functions of this model. The OSI model is a system that helps you break down network functions. For example, something we networkers hear all the time is: Oh! The network is down, the internet is down, or I am not able to access a server. That’s when you need to focus to understand what they are trying to describe, because network is a big place where lot of things could go wrong. Thus the OSI model helps us to break down the network functions and resolve potential issues quickly.

There are vendors that make specific devices for specific layers. I know I haven’t described the layers yet but let me give you a quick example without going ahead of myself. The Physical layer describes physical connectivity. When you think of physical layer, literally think of physical things that you can hold in your hands like network cards, cables or WAN interface cards. Some manufacturer’s focus is nothing but physical layer equipment. The OSI model creates standards for those equipment’s manufacturers and allow vendors to specialize in specific areas of the network equipment.

Understanding The Layers of OSI

Application: I took the liberty to start from the top of the OSI model which is the application layer. This layer is a very familiar layer because it’s usually the one where we deal with the application itself. It provides an interface that allows applications to communicate across the network. If you are using some kind of application that is ‘network aware’ and has the ability to send data across the network like Web Browser, an e-mail application, or even an online game that communicates on the network, this is the point where it begins to communicate.

Presentation: Now as soon as the application starts sending data, it passes through the presentation layer. This layer “generi-fies” the data, meaning the data becomes formatted in a way that is understandable by whatever server you are communicating with. For example, you go to www. itmotion.net. This webpage is formatted in html format, which is a format most web pages understand. On the same page you might have pictures that are in jpeg, which is a format every webserver and every web client understands. That is the function of the presentation layer, it presents the data compliant to standards so when they talk to each other, they understand each other. There is also encryption services on the presentation layer, namely generic encryption services, which is something the entire internet uses. A good example is when you go checking your online banking account balance, you expect your request to be secure and encrypted. and It is in a generic way; it uses a set of keys that both the client and the server possess to encrypt the data back and forth. And it is very secure.

Session: The data is now generi-fied and passed to the session layer. This layer starts and ends a session. When you first access a web server, it goes “hey I am ready to begin a session; let me communicate with you”, and it keeps all the sessions separate. For example, when you surf the internet and open numerous applications at the same time, you have to have some kind of system that allows the right data to go to the right program.

As we move down to the transport layer, I want to draw a line. In Cisco’s world, the above three layers are the least important layers because the application, the presentation, and the session happen before the data leaves your computer. In other words, all the above three layers are handled by Windows, Linux or whatever Operating system you run on computer.

Transport: The important layers start from Transport layer, to which I would award the certificate of most important layer of all because it is a layer that dictates how the data is sent. When the application get to the transport layer, it either has to send the data across the network reliably or unreliably (TCP or UDP, respectively). To read more about TCP operations click here. If the application choses a reliable transport when it sends a packet to the server, it will expect that server to send another packet back called ACK ,an acknowledgement meaning I got it. Thus if it chooses reliable, then every single thing I send needs to be acknowledged so the application knows it is received; if it doesn’t get the ACK, it will send the packet again. With that definition, you must be thinking why on earth will the application send packet unreliably? Unreliable transport is typically used for something known as Real-time application. For example, one of the hot new things is voice over ip (VoIP), meaning I have phones that run over the network and as I speak in to the phone it is sending my voice packets over the network, or Video over ip where you actually watching television online or stream a movie. The point in this is the packets are going in real time; if it drops a packet and you are on your phone talking to your mom, she will ask you to repeat the words. The packets don’t care about the acknowledgement because everything is happening in real time.

The other function of the transport layer is describing what is known as Port Numbers. What port numbers allow you to do is designate what service you trying to access. For example, when you want to surf the web from your home computer, the web application (IE or Firefox) has to go to the web server and grab the page. The server needs to know that the client is asking for the webpage. That server may be running other services; the port number is how the server knows it is being asked for the webpage and not a different service.

Network: This layer provides something know as logical addressing, which most people referred to as IP addressing. When you assign an ip address to your computer or server, it is happening at the network layer. This layer also finds the best path to a destination. When you go to www.itmotion.net, your request has to go around the network through the internet to right here in Washington DC, which is where ITmotion web server resides. From wherever you are located, the routers have to find the best path to get to their destination.

Datalink layer: Now you have something known as Physical addressing, which most people call the MAC address. This an address that is planted on a network card. It is mainly used for communicating just on your network, meaning it allows your computer to speak to the computer down the hall or a server that is on the same network. Datalink layer also ensures that the data is error –free.

Physical: This layer is where we have physical access to the cable. You move from all these logical concepts down in to electric signals ones and zeros. When you think of physical layer, literally think of physical things that you can hold in your hands like network cards, cables, WAN interface cards.