OSI – Standardization within Computer Networks
A network is a complex structure within computers. Here, all important data transmissions are gathered, and it represents all connections for internal device communication and coupling to the control unit. As this connection describes complex processes, protocols must be used. Protocols make it possible to control the communication in a system in order to assure that the data of any formatting is directed to the addressee at the time he needs to receive them.
Protocols also control the accuracy of the transmitted messages and information. This means that the contents which are supposed to be in a data packet actually are in the packet. As there are several technologies in this area, standards and norms are being specified continuously in order to obtain an internationally recognized standard. The OSI model, also known as the OSI layer model or the OSI reference model, is one of them.
Structure of the OSI Model
The development of this model began in 1979, and four years later it was specified as a standard. For regulation of tasks which occur with communication between the participants of a network, they are divided into layers, and the OSI model has a total of seven layers.
Each layer is occupied by an instance, where the instances must operate according to prescribed rules on the side of the sender as well as the receiver, in order to come to an agreement about the method of data processing. This regulation also claims to be a logical, horizontal connection. However, this only applies to the instances within a layer, as the data flow between a higher and a lower layer runs vertically.
The seven layers of the OSI model build upon each other, but they differ in the way that they realize their requirements in the form of a communication protocol. Over time, this had the consequence that each individual layer by now has several protocol types available.
This layer in the OSI model serves for bit transfer. It makes mechanical as well as electrical resources available, which serve for activation and deactivation of physical connections, as well as maintenance of these lines and the transfer of bits. Within this layer, bits are sent on a line-bound transfer path or on one without a line. Finally, the requirements also include the regulation of the transfer method, namely which information carrier is selected.
This is the security layer, which can also be called the section security layer, the connection security layer, or the data security layer. After all, this layer in the OSI model serves for exactly this, assuring a secure and as far as possible error-free transfer by securing connections, data, and sections. Furthermore, this instance also performs regulation of the access to the transmission medium. Additionally, checksums are added here, a process taking place within the frame of a channel coding. The purpose of this measure is to find faulty blocks and enabling them to be dropped by the receiver. These blocks can be corrected when required, but once dropped blocks cannot be requested again in this layer.
This is the network layer, and it takes care of the switching of connections within performance-oriented services and transferring of data packets with packet-oriented services. as the transmission of the information always covers the entire communication network, finding routes between the nodes of a network is also one of the tasks of this layer. More precisely the main tasks of the third layer of the OSI model provide addresses via networks, update routing tables and create them, negotiate and assure a special service quality, and fragment the data packets to be transferred.
This is the transport layer of the OSI model, which enables uniform access for the superior layers and takes care of segmenting of the data flow and avoiding of congestion. Five different service qualities are defined within the layer, which can be used by the layers 5 to 7 and which contain services with multiplex mechanisms and error correction methods.
This layer is the session layer, serving for communication of the processes of two systems and making available services of the organized and synchronized data exchange. These avoid a crash of the session. To fulfill these tasks, check points, i.e. fixed points. are specified, so that new synchronization can be performed from this position in case of a failure of the transports.
The presentation layer enables syntactically correct flow of the data exchange by selecting a representation of the data that is independent of system and form. Data compression, encryption, and assuring the readability of data by a different system are also part of this instance. When required, the presentation layer of the OSI model also assumes the position of a translator between two systems.
This is the application layer and also the top layer in the OSI model. Access to network services takes place here.
Conclusion for the OSI Model
Without this model, realization of various processes within the communication technology would not be possible, as it is the base for the development of numerous protocol types and still serves as a design model for uniform protocols and thus is indispensable for the order of a network.
THE OSI MODEL – PART II
OSI model is the designation for a model for the functional stratification of communication protocols in computer networks. OSI is the abbreviation of Open Systems Interconnection reference model, often also called ISO/OSI. This designation makes reference to the fact that the model has beens standardized by the International Organization for Standardization.
The OSI model provides for a structure of seven layers. Here, every layer except the top layer provides functions for the layer above it. With this, each layer only has to know the interface definition of the next layer down, but not the concrete implementation of this lower layer or the layers below this.
1st Layer: Bit transmission
The lowest layer of the OSI model is called the bit transmission layer. The designation as a physical layer clearly indicates what actually happens on this layer: Here, it is a matter of actual transmission of individual bits and how this transmission actually occurs on the physical medium. While the implementation of this layer is quite simple for classic electrical conductors as the medium, for other media like glass fiber or for transmission via electromagnetic waves (W-LAN), they must be capable of representing bits or packets of bits for example as specific wavelengths. Together with the second layer, this layer is the only layer where the actual nature of the medium must be taken into consideration.
2nd Layer: Security layer
The security layer, also called the link layer, assures the consistency of the transmission. Amongst others, this layer provides the transmitted bit packets (frames) with checksums, which enable the receiver to check whether a frame has been transmitted correctly or whether an error has occurred during the transmission. Depending on the type of the error, it is either possible to correct the frame itself or to request the sender to repeat the transmission. The sending of receipt confirmations also takes place on this layer. If the sender does not receive such feedback from the receiver, it assumes that a transmission error has occurred and repeats the transmission.
3rd Layer: Network layer
The network layer of the OSI model is in charge of routing in the network. While with packet-based services, for example TCP/IP, a path is searched for each bit of data (packet), which may differ strongly from packet to packet, for line-based services a unique path is selected between the two communication partners, and it remains the same during the same transmission, unless a line failure or a similar error occurs. A widely distributed protocol on this layer is the Internet Protocol (IP), which should be known even to less versed users because of the IP address assigned to each device in an IP-based network, like for example the Internet.
4th Layer: Transport layer
The transport layer of the OSI model combines the individual packets again into the so-called segments. A cross-packet error correction may also take place in this layer, i.e. further errors are detected and corrected if possible or a repeat sending is requested from the sender.
5th Layer: Session Layer
In the OSI model, the session layer is in charge of establishing a session between the two communication partners and to manage it. This also includes the re-establishment of any interrupted communication and to continue at the point where it was interrupted.
6th Layer: Presentation layer
The tasks of the presentation layer include the offering of cross-protocol functions. Accordingly, with text-based protocols, individual bytes again become a readable text in this layer, or in the case of transmission with special protocols for video transmission, images and similar are processed. A cross-system processing of umlaut characters can also be done in this layer.
7th Layer: Application layer
The top layer is the application layer, which interacts directly with the end-user program and is finally operated by the user. Considering the World Wide Web as a very frequently used service, then it is the Hypertext Transport Protocol (HTTP), used amongst others in the application layer, that interacts directly with the browser. This again is operated by the user.
TCP/IP Protocol Stack as Application of the OSI Model
The Internet is based on the TCP/IP protocol stack. Here, the first two layers of the OSI model are combined to form the network access layer. Protocols like Ethernet or IEEE 802.11 (for WLAN) are used in these layers. The naming protocols IP and TCP are used on the network or the transport layer. The top three layers of the OSI model are finally combined to the application layer of the TCP/IP protocol stack.
Advantages of the Model
The advantages of the OSI model are obvious on closer inspection. Thus, the division of the communication into individual layers stacking onto each other has the advantage that it is possible, for example to change the physical medium without having to change the protocol layers above the data link layer. This modularization also makes it easier to determine that implementation is correct, as the layers can first be checked by themselves, so that any errors can already be corrected.