Introduction Internet Protocol Suite Essay Example

services is a way to abstract upper layer protocols from being detected, while the lower layers are spared from needing to understand the intricacies of each individual application and its protocol.

The architectural documents related to layers, unlike ISO 7498 or the Open Systems Interconnection (OSI) model, have fewer and less precisely defined areas than the OSI model. This makes them more compatible with real-world protocols. In contrast, a commonly referenced document, RFC 1958, does not feature a layered stack. The emphasis on layering is a significant distinction between the approaches of IETF and OSI.

The text discusses the presence of an "intertwining layer" and "upper layers" in the internet architecture. It highlights that the architecture has evolved over time and does not adhere to a specific plan. However, it is necessary to document the current principles of the internet architecture in 1996. RFC 1122, titled Host Requirements, outlines these principles and refers to various other architectural principles beyond layering. It presents a loosely defined four-layer model with named layers instead of numbered ones.

The first layer, known as the application layer, deals with communication between applications on the same or different hosts. This layer includes protocols such as ESMTP, FTP, SSH, HTTP, etc.

The second layer is referred to as the transport layer and facilitates networking between two network hosts, whether they are on a local network or separated by routers. It provides a uniform interface that hides the underlying network connections' topology.

This layer encompasses various functions like flow-control, error-correction, and connection protocols like TCP. Its responsibility lies in establishing and managing connections between internet hosts.

Additionally, the internet layer enables datagram exchange

across different network boundaries.

The layer that establishes intertwining and defines the Internet and its addressing and routing structures for the TCP/IP protocol suite is also known as the primary protocol layer. It is commonly referred to as the Internet Protocol (IP) layer, which defines IP addresses. This layer ensures that datagrams are transported to the next IP router connected to a network closer to the intended data destination. On the other hand, the link layer defines networking methods used by hosts to communicate within a local network link without the involvement of routers.

The local network topology and interface protocols for transmitting Internet layer datagrams to neighboring hosts are described in this layer (CB. The OSI data link layer). The TCP/IP model and layered protocol stack design were already in use prior to the establishment of the OSI model. The comparison between the TCP/IP model and OSI model in various resources like books and letters often causes confusion because they have different assumptions about the importance of strict layering.

The abstraction enables upper layers to offer services that the lower layers cannot or decide not to offer. In addition, the original OSI model was expanded to include a best effort delivery protocol. This implies that transport layer implementations have to decide whether or not to provide reliability and to what extent. JODI ensures data integrity through a checksum but does not ensure delivery. On the other hand, TCP provides both data integrity and delivery guarantee by retransmitting until the packet is acknowledged by the receiver.

This model does not have the formalism of the OSI model and its associated documents. However, the IETF does

not use a formal model and does not see this as a limitation, as stated by David D. Clark, "We reject: kings, presidents and voting. We believe in: rough consensus and running code." Criticisms of this model, similar to those made about the OSI model, often do not take into account Sis's later extensions to that model. 1. To establish connections between MultiMate and their own addressing systems (e.g. Ethernet), an address mapping protocol is required.

The protocols discussed can be classified as being below IP but above the existing link system. According to an extension to the OSI model called MILLION, this can be regarded as a subsection dependent convergence facility within the internal organization of the network layer. IGMP and IGMP exist on top of IP but do not transport data like JODI or TCP. These functions are considered as layer management extensions to the OSI model within its Management Framework (OSIER MFC). The SSL/TLS library operates above the transport layer, using TCP, but below application protocols.

Once again, the designers of these protocols had no intention to adhere to the OSI architecture. Moreover, the link is considered as a black box in this context, which is suitable for discussing IP as it can operate over any medium. Notably, the IETF does not aim to address transmission systems unlike the OSI model. Below, the description of each layer in the TCP/IP networking model commences from the lowest level - the link layer. Here, the link layer refers to the networking scope of the local network connection to which a host is connected.

This regime is known as the link in Internet literature. It

is considered the lowest layer of the Internet protocols because TCP/IP is designed to be independent of any specific hardware. Therefore, TCP/IP can work with almost any type of hardware networking technology. The link layer's purpose is to facilitate the movement of packets between the Internet layer interfaces of two hosts on the same link. The transmission and reception of packets on a specific link can be controlled through both the software device driver for the network card and firmware or specialized chips.

The data link layer in the TCP/IP model has various functions, including adding a packet header and transmitting the frame over a physical medium. It also includes specifications for translating network addressing methods to data link addressing, such as Media Access Control (MAC). However, all aspects below this level are assumed to exist without explicit definition. Additionally, this layer is responsible for selecting packets to be sent over a virtual private network or other networking tunnel.

The link layer data in this scenario can be seen as application data that is transmitted or received over another IP connection, known as a virtual link. This connection is established with the link layer of the protocol stack. In the TCP/IP model, there is no strict hierarchical encapsulation sequence. The internet layer enables packet transmission across multiple networks by sending data from the source network to the destination network.

Routing is the act of sending data packets from a source to a destination by directing them to the next network node, called a router. This action is facilitated by the Internet Protocol (IP) in the Internet protocol suite, which also handles host addressing and identification

through a hierarchical addressing system. The internet layer within IP does not differentiate between different transport layer protocols and has no awareness of application data structures.

The IP protocol can transmit data for different upper layer protocols, each with its own unique protocol number. For instance, IGMP and IGMP have protocol numbers 1 and 2 respectively. Although some protocols transmitted by IP, such as IGMP for diagnostic information transmission and IGMP for IP Multicast data management, are layered on top of IP, they still have interconnected functions. This highlights the contrasting architectures between the Internet's TCP/IP stack and the OSI model.

The internet layer's main function is to enable the networking of different IP networks by forwarding transport layer datagrams to next-hop routers. This facilitates the intertwining of networks and establishes the Internet. The Internet Protocol is the core element of the internet layer and it defines two addressing systems for identifying and locating network hosts.

The ARPANET and the Internet originally used a 32-bit IP address system called Internet Protocol version 4 (IPPP) that could identify about four billion hosts. However, in 1998, this system's limitations were overcome with the introduction of Internet Protocol version 6 (IPPP), which started being implemented around 2006. The transport layer is responsible for establishing connectivity between hosts and managing data transmission, regardless of the structure or purpose of the user data exchanged.

The transport layer is responsible for end-to-end message transfer regardless of the underlying network, and it handles tasks such as error control, segmentation, flow control, congestion control, and application addressing (port numbers). The transport layer can be classified as either connection-oriented (TCP) or connectionless (UDP), depending

on the type of message transmission or application connection. It acts as a transport mechanism, similar to a vehicle, ensuring the safe delivery of its contents (passengers/goods) to the intended destination, unless another protocol layer is responsible for safe delivery. The layer establishes a basic data channel that an application can utilize for specific data exchange purposes. To enable this, the layer establishes communication channels that an application requires. In order to address specific services offered by a server computer without the need for service announcements or directory services, standardized port numbers are available for many types of services.

The transport layer is the first layer of the TCP/IP stack to offer reliability because IP only provides a best effort delivery. IP can be used with a reliable data link protocol like High-Level Data Link Control (HDLC). TCP, for example, is a connection-oriented protocol that addresses multiple reliability issues to ensure a reliable byte stream: data arrives in order, has minimal error (i.e. correctness), duplicate data is discarded, lost/discarded packets are resent, and traffic congestion control is included. Another reliable, connection-oriented transport mechanism is the newer Stream Control Transmission Protocol (SCTP).

The message-stream-oriented protocol, unlike TCP, supports multiple streams multiplexed over a single connection. Additionally, it allows for multi-homing support, where a connection end can be represented by multiple IP addresses corresponding to various physical interfaces. This ensures uninterrupted connection even if one of the interfaces fails. Originally designed for telephony applications to transport ASS over 'P, this protocol can also be utilized for other applications. User Datagram Protocol (UDP) is a connectionless datagram protocol that, similar to 'P, operates as a best effort and

"unreliable" protocol.

The significance of reliability in various protocols is discussed in the text. A less powerful checksum algorithm is employed for error detection. UDP (User Datagram Protocol) finds common usage in applications like streaming media, where timely arrival takes precedence over reliability. It is also utilized in uncomplicated query/response applications such as DNS lookups, where establishing a dependable connection incurs significant overheads. In contrast, Real-time Transport Protocol (RTP) is specifically designed for real-time data like streaming audio and video. The TCP or UDP port distinguishes between applications at a particular network address.

In general, well-known ports are frequently linked to specific applications. (See List of TCP and UDP port numbers.) The application layer consists of higher-level protocols that most network communication or applications rely on. These protocols encompass the File Transfer Protocol (FTP) and the Simple Mail Transfer Protocol (ESMTP). The data encoded with application layer protocols is then encapsulated within one or more transport layer protocols, such as TCP or UDP, which utilize lower-layer protocols to facilitate the actual data transfer.

The absence of layers between the application and transport layers in the IP stack requires the application layer to incorporate protocols that function similarly to the presentation and session layer protocols of the OSI model. This is typically achieved using libraries. The application layer protocols typically view the transport layer and lower protocols as opaque entities that offer a reliable network connection for communication. However, these applications are generally aware of important attributes of the transport layer connection, such as the IP addresses and port numbers of the endpoints.

In the Internet protocol suite, layers are not clearly defined. Application layer protocols

are typically assigned ports by the IANA, such as HTTP with port 80 and Telnet with port 23. Clients, on the other hand, use ephemeral ports, which are port numbers assigned from a specific range. Transport and lower level layers do not focus on the details of application layer protocols. Routers and switches simply provide a conduit for the encapsulated traffic without inspecting the specific application protocol being used.

However, certain firewall and bandwidth throttling applications, such as the Reservation Protocol (RSVP), do attempt to analyze the content. Additionally, Network Address Translation (NAT) utilities may need to consider the requirements of specific application layer protocols. NAT enables devices on private networks to communicate with the external network through a single visible IP address using port forwarding, and is commonly found in contemporary household broadband routers.

The differences between OSI and TCP/IP layering can primarily be observed in the upper layers. The OSI model comprises three distinct layers - the application layer, the presentation layer, and the session layer. In contrast, the TCP/IP model consolidates these three layers into a single application layer. While certain pure OSI protocol applications, such as X.400, also integrate these layers together, there is no requirement for a TCP/IP protocol stack to possess a unified architecture above the transport layer. For example, the NFG application protocol operates over the external Data Representation (CDR) presentation protocol which is then implemented utilizing Remote Procedure Call (RPC) as its underlying protocol.

RPC enables secure record transmission over UDP transport and is subject to varying interpretations of the RFC regarding the coverage of OSI model layer 1 and the assumption of a hardware layer below

the link layer in the TCP/IP model. Some authors have tried to integrate layers 1 and 2 from the OSI model into the TCP/IP model to align with modern standards. This can result in a five-layer model, with the link layer or network access layer divided into layers 1 and 2 of the OSI model.

The session layer is analogous to the Telnet virtual terminal functionality found in text-based protocols like HTTP and ESMTP in the TCP/IP model's application layer protocols. Additionally, it aligns with TCP and JODI port numbering, categorized as part of the transport layer in the TCP/IP model. Certain tasks typically handled by an OSI restoration layer are conducted at the Internet application layer through the utilization of the MIME standard, which is employed in application layer protocols like HTTP and ESMTP.