Internetworking

The Internet is a term with many meanings, depending on the context of its use . To the general public, the term is often used synonymously with the World Wide Web, its best-known application . But the internet supports many other applications, such as electronic mail, streaming media, such as internet radio and video, a large percentage of telephone traffic, system monitoring and real-time control applications, to name a few.

In one respect the Internet is similar to an iceberg. The vast majority of it is out of sight. While these distributed applications allow users to utilize internet services, in the context of convergence of communications, they require a large suite of technologies visible only to the enterprises that provide them. To Internet Service Providers, the Internet identifies these underlying services. There are internet services that are accessible to the general public, while the same technologies providing similar services are available in restricted environments, such as those in an enterprise intranet, in military and government private internets and in local home networks. Further complicating the notion of an Internet is is the frequent interconnection of public and private networks in ways that allow limited interaction. This article and the subgroup it describes uses the term Internet in the broadest sense. That is, it identifies the applications that provide an interface between users and communications services, those services themselves, public and private instances of application and communications services and the aggregation of private and public networks into a global communications and application resource.

The architecture of the Internet

The development of the internet shows it as the culmination of significant activity in both the commercial world as well as within government sponsored programs. While the main development occurred in the United States, there were major contributions from researchers and engineers in the U.K., France and other parts of Europe. This work led to the existing architectural model.

In order to engineer the internet, internet designers and engineers place its services into one of several layers, which in total comprise the internet protocol architecture. Internet architectural experts deprecate an overemphasis on layering; the more important principles of Internet architecture include:

• End-to-End Principle: Application intelligence is at the edge of the cloud; there have been variations on this principle.
• Robustness principle: "Be conservative in what you send, be liberal in what you receive."

While there have been several different protocol architecture designs, the one with the strongest support consists of 4 layers: 1) the application layer, 2) the transport layer, 3) the internet layer, and 4) the link-layer.. Each protocol layer utilizes the services of the next lower layer (except the lowest, the link layer) to provide a value-added service to the layer above it (except for the application layer, which provides services to users). Utilizing this protocol architecture, it is possible to describe how the Internet works.

Web browsers are the most common user interface in the Internet. Such browsers translate human requests to the Hypertext Transfer Protocol (HTTP), which actually moves data between the browser and a Web server. Consequently, measured solely in terms of percentage of use, the World Wide Web is the most frequently used Internet application. (However, this is expected to change. Forecasts of Internet bandwidth utilization suggest that video traffic will make up over 90% of Internet traffic by 2013. ). The communications services provided by the Internet have no direct human interfaces; every user-visible function must go through a program resident on a client or server computer. There are literally hundreds of different protocols, applications and services that run over the Internet. Virtual private networks interconnecting the parts of individual enterprises, or sets of cooperating enterprises, overlay the Internet. As mentioned previously a wide range of interconnected networks using the same protocols as the public Internet, but isolated from it, provide services ranging from passing orders to launch nuclear weapons, authorizing credit card purchases, collecting intelligence information, controlling the electric power grid (see System Control And Data Acquisition), telemedicine such as transferring medical images and even allowing remote surgery, etc. Many of these applications utilize custom application interfaces that do not involve a web browser. Consequently, internet distributed applications comprise a much larger set than those visible to the general public.

In addition to applications that are directly experienced by Internet customers, there are a wide-range of internet applications that exist to provide infrastructure services to the internet. Examples of infrastructure services are the Doman Name System (DNS), which associates computers connected to the Internet with human friendly names. The movement of data through the internet requires that it visit intermediate systems called routers. The activity of directing the data through the internet, called routing, utilizes an infrastructure application that distributes routing data to routers. The secure identification of users to applications requires the use of authentication servers, such as RADIUS and Kerberos, each of which is a distributed application in and of itself. These are just a few of the internet infrastructure applications that support the provision of internet service.

Internet applications are distributed. That is, they normally are comprised of components that reside at different locations. That means they must exchange data through communications equipment that is subject to various failure modes. Furthermore, one element may have the capability to send data faster than the receiver can process. The next layer in the protocol architecture, the transport layer, provides services that address these issues. Transport layer protocols, like the Transmission Control Protocol (TCP) provide end-to-end error management and flow-control services that ensure application elements can exchange data in an error-tolerant and synchronized manner. Instead of relying on the error and flow-control services provided by TCP, some applications handle these services themselves. Those that do, utilize a datagram service also provided by the transport layer. For example the Unreliable Datagram Protocol (UDP) moves packets between application parts without the provision of either error-control or flow-control services.

The next layer of internet service, the internet layer moves data between end-systems (normally customer computers, but in some cases infrastructure systems) through an interconnected set of systems, called routers, which are mentioned above. Routers come in all shapes and sizes. Some, normally located at the periphery of the internet such as those in a home or small business, are known as edge routers. Others are service provider equipment with varying capabilities, from modest performance border routers to high performance core routers. These routers are interconnected, moving data across the Internet in a way that increases the probability of successful transit. There are two types of routing schemes. Virtual circuit routing reserves resources over a fixed path between two end-systems. Packet routing operates in a way whereby individual packets of data may take different paths through the systems that interconnect end-systems. The internet layer also supports specialized data services, such as multicast, broadcast, and anycast routing.

Routers and end systems connect to each other through the [lLink layer]]. This layer may comprise a physical channel or a complex networking infrastructure. Both are commonly deployed options.

Physical channels encode data utilizing various techniques, thereby providing the basic data transmission service between directly connected equipment. There are a wide variety of physical channels, each utilizing its own data encoding scheme. Examples of physical channels used in the Internet are wire-based channels, such as those used by low-bandwidth ethernet; wireless broadcast channels, such as those used in Wi-Fi, also known as 802.11, as well as in cell phone service; optical channels, such as those used by high-bandwidth ethernet; and wireless point-to-point radio channels, such as those used by microwave links and satellite communications. Since physical channels may introduce communications errors and generally do not provide flow control, the link-layer may provide services that correct most errors and also implement flow control. The characteristics of the physical channel may vary widely from the fairly reliable ethernet, less reliable wireless channels, to the very unreliable deep space radio channels. Consequently, each type of physical channel may require a different link-layer protocol to accommodate its characteristics. For example, normally ethernet channels provide only forward error correction and no flow control services. Low to moderate data rate serial channels, on the other hand, may provide acknowledgment based error and flow control.

When the link layer comprises networking infrastructure, it implements a technique known as network overlaying. This scheme encapsulates the packets of the internet layer inside packets of the link layer network. Common examples are carrying internet traffic over an ATM network, which is a virtual circuit communications network. Sometimes it is useful to encapsulate internet packets inside other internet packets. For example, a private intranet may wish to interconnect several isolated sites using the services of the public internet. It protects its internet packets with a suitable security protocol, such as IPSEC and places them inside the internet packets of the public network, which moves them between these isolated sites.

The Internet utilizes not only technology acting within a single layer of its protocol architecture, but also mechanisms that are spread over several protocol layers. As mentioned previously, routing is one such technology using application services to move routing data to routers in order to provide the network-layer routing service. Another example is the provision of network security within the Internet. For example, providing a secure transport service requires encrypting of packets at end-systems This requires encryption keys that are distributed by a logically separate application. Internet management may utilize an application layer protocol, such as the Simple Network Management Protocol (SNMP) in concert with a network-layer protocol, such as the Internet Message Control Protocol (ICMP).

Professional societies and organizations

(See External Links subpage for website homepages)

• International: Internet Society (ISOC), IEEE Communications Society (IEEE ComSoc), World Wide Web Consortium (W3C), Internet Technical Committee (ITC), Association for Computer Machinery Special Interest Group on Data Communications (ACM SIGCOMM), Internet Corporation for Assigned Names and Numbers (ICANN); International Telecommunications Union (ITU), International Electrotechnical Commission (IEC).
• North America: North American Network Operators Group (NANOG), American Registry for Internet Numbers (ARIN)
• Europe: European Telecommunications Standards Institute (ETSI), Réseaux IP Européens (RIPE), RIPE Network Coordination Centre (RIPE-NCC)
• Asia: Asia-Pacific Network Information Center (APNIC), South Asian Network Operators Group (SANOG)
• Middle East: Middle East Network Operators Group (MENOG)
• Africa: African Network Operators Group (AfrNOG)
• Pacific: The Pacific Network Operators Group (PacNOG)
• Latin America: Latin America and Caribbean Network Information Center (LACNIC); Latin America and Caribbean Region Network Operators Group (LACNOG)
• France: FRench Network Operators Group (FRnOG)
• United States: Telecommunications Industry Association (TIA)