 The importance of understanding how the Internet is laid out is essential to understanding how the next generation networks are going to be like. In this particular module, we'll see how the International Standards Organization has developed an open system interconnect model to describe the architecture of any kind of network. Then specifically, we'd look at how the Internet is organized. The ISO has defined that telecommunication networks can be characterized by a distributed architecture from different perspectives. For instance, how the protocols interact with each other, how the network is shaped up, and how the services are provided to the users. The seven-layer ISO OSI stack, shown to you on the left-hand side, is quite general in nature. It can be translated to any kind of network that we are intending to. In the case of Internet, as we can see that starting from layer 5 to 7 of the ISO OSI are grouped into the application layer, the transport layer conforms to the transport layer as such, and the network layer, also called the IP layer, corresponds to the network layer of the seven-layer architecture. These upper layers can also be considered to be the software layers which implement the protocols that are used for communication over the Internet. The lower two layers, that is the data link layer and the physical layer, are realized mostly in hardware and are responsible for providing a platform onto which the upper layers are executed. Internet can also be understood as a topology, a graph that is based on certain entities. These entities in the Internet jargon are the end hosts which implement all seven layers of the ISO OSI stack, a repeater which only implements physical layer, the hub which also implements only the physical layer, but it is a multi-port repeater that allows the triple R operation for multiple hosts. Then we have the bridge. These bridges are used to isolate different networks so that the collision domains of these networks do not overlap. A multi-port bridge is also known as a switch, and lastly we have a layer three device that is a router which isolates networks on the basis of their IP addressing schemes and allows the organization of each network as an independent entity administered by an independent administrator. If you look at this figure, it looks more like a tree where the higher order or the ancestor node is grouped to the descendant nodes which are also its children. So we can see that starting from local area networks which converge into a metropolitan area network and multiple metropolitan area networks can then feed into a wide area network. These WANs are in turn connected to each other as autonomous entities through a gateway router. We can also appreciate the presence of certain interesting devices like a Wi-Fi hotspot shown to you at the bottom and a smart device such as a tablet or an iPad are also shown to you next to it. So what it implies is that this internet topology comprises both wide and wireless architecture. There are certain interesting aspects to the internet architecture. The first and foremost is that internet emerged as a best effort network. Of course it can attempt to provide value added services but to begin with essentially it is best effort. Because of that it allows scalability because an arbitrary number of users can be multiplexed together. And the type and nature of these devices could vary which means that the internet architecture is heterogeneous. Please note that these are the pluses. In fact these are the strengths of the internet architecture that helps it to eventually emerge as the dominating candidate for next generation networks. Internet architecture is also increasingly flexible. Starting from the initially perceived client server architecture we have seen the torrents in the form of peer-to-peer networks and the resultant service oriented architecture in which every service provider and every service solicitor are connected through the internet architecture on the fly.