OSI stack

There is a clear distinction to be made between the OSI model and the OSI stack. While the OSI model is a conceptual blueprint for how open systems interconnect, the OSI stack is a set of very specific protocol specifications. Unlike other protocol stacks, the OSI stack fully follows the OSI model and includes protocol specifications for all seven interoperability layers defined in the model. At lower levels, the OSI stack supports Ethernet, Token Ring, FDDI, protocols global networks, X.25 and ISDN - that is, it uses lower-layer protocols developed outside the stack, like all other stacks. The protocols of the network, transport and session layers of the OSI stack are specified and implemented by various manufacturers, but are not yet widespread. The most popular protocols in the OSI stack are application protocols. These include: FTAM file transfer protocol, VTP terminal emulation protocol, X.500 help desk protocols, X.400 email protocols and a number of others.

The protocols of the OSI stack are characterized by great complexity and ambiguity of specifications. These properties were the result of the general policy of the stack developers, who sought to take into account all use cases and all existing and emerging technologies in their protocols. To this we must also add the consequences of a large number of political compromises that are inevitable when adopting international standards on such a pressing issue as the construction of open computer networks.

Due to their complexity, OSI protocols are expensive computing power CPU, making them most suitable for powerful machines rather than networks personal computers.

The OSI stack is an international, vendor-independent standard. It is supported by the US government in its GOSIP program, according to which all computer networks U.S. government installations after 1990 must either directly support the OSI stack or provide a means to migrate to that stack in the future. However, the OSI stack is more popular in Europe than in the US because there are fewer legacy networks left in Europe running their own protocols. Most organizations are still planning their migration to the OSI stack, and very few have started pilot projects. Among those working in this direction are the US Navy Department and the NFSNET network. One of largest producers AT&T is a company that supports OSI; its Stargroup network is entirely based on this stack.

TCP/IP stack

The TCP/IP stack was developed at the initiative of the US Department of Defense more than 20 years ago to connect the experimental ARPAnet network with other networks as a set common protocols for heterogeneous computing environments. Berkeley University made a major contribution to the development of the TCP/IP stack, which got its name from the popular IP and TCP protocols, by implementing the stack protocols in its version of the UNIX OS. The popularity of this operating system led to the widespread adoption of TCP, IP and other protocol stacks. Today, this stack is used to connect computers on the Internet, as well as in a huge number of corporate networks.

The TCP/IP stack at the lower level supports all popular standards of the physical and data link layers: for local networks- these are Ethernet, Token Ring, FDDI, for global - protocols for working on analog switched and dedicated SLIP lines, RRR, protocols territorial networks X.25 and ISDN.

The main protocols of the stack, which give it its name, are IP and TCP. These protocols, in OSI model terminology, belong to the network and transport layers, respectively. IP ensures that the packet travels across the composite network, and TCP ensures the reliability of its delivery.

Over many years of use in the networks of various countries and organizations, the TCP/IP stack has incorporated a large number of application level protocols. These include such popular protocols as the FTP file transfer protocol, the telnet terminal emulation protocol, the SMTP mail protocol used in Internet e-mail, hypertext services of the WWW service and many others.

Today, the TCP/IP stack is one of the most common transport protocol stacks in computer networks. Indeed, the Internet alone connects about 10 million computers around the world that interact with each other using the TCP/IP protocol stack.

The rapid growth in the popularity of the Internet has also led to changes in the balance of power in the world of communication protocols - the TCP/IP protocols on which the Internet is built began to quickly push aside the undisputed leader of past years - Novell's IPX/SPX stack. Today in the world the total number of computers on which the TCP/IP stack is installed has become equal to the total number of computers on which the IPX/SPX stack is running, and this indicates a sharp change in the attitude of local network administrators to the protocols used on desktop computers, since They make up the overwhelming majority of the world's computer fleet, and it was on them that Novell's protocols, necessary for access to NetWare file servers, used to work almost everywhere. The process of establishing the TCP/IP stack as the number one stack in any type of network continues, and now any industrial operating system necessarily includes a software implementation of this stack in its delivery package.

Although the TCP/IP protocols are inextricably linked with the Internet and each of the multimillion-dollar armada of Internet computers runs on the basis of this stack, there are a large number of local, corporate and territorial networks that are not directly parts of the Internet that also use TCPDR protocols. To distinguish them from the Internet, these networks are called TCP/IP networks or simply IP networks.

Because the TCP/IP stack was originally designed for the global Internet, it has many features that give it an advantage over other protocols when it comes to building networks that include wide-area communications. In particular, a very useful property that makes possible use of this protocol in large networks is its ability to fragment packets. Indeed, a large composite network often consists of networks built on completely different principles. Each of these networks can set its own value for the maximum length of a unit of transmitted data (frame). In this case, when moving from one network with a larger maximum length to a network with a smaller maximum length It may be necessary to divide the transmitted frame into several parts. The IP protocol of the TCP/IP stack effectively solves this problem.

Another feature of TCP/IP technology is its flexible addressing system, which makes it easier to include networks of other technologies into the Internet compared to other protocols of similar purposes. This property also facilitates the use of the TCP/IP stack for building large heterogeneous networks.

The TCP/IP stack uses broadcast capabilities very sparingly. This property is absolutely necessary when working on slow communication channels characteristic of territorial networks.

However, as always, you have to pay for the benefits you get, and the price here is high resource requirements and the complexity of administering IP networks. The powerful functionality of the TCP/IP protocol stack requires high computational costs to implement. A flexible addressing system and the elimination of broadcasts lead to the presence of various centralized services in the IP network DNS type, DHCP, etc. Each of these services is aimed at facilitating network administration, including facilitating equipment configuration, but at the same time, it itself requires close attention from administrators.

There are other arguments for and against the Internet protocol stack, but the fact remains that today it is the most popular protocol stack, widely used in both global and local networks.

IPX/SPX stack

This stack is the original Novell protocol stack, developed for the NetWare network operating system back in the early 80s. The network and session layer protocols Internetwork Packet Exchange (IPX) and Sequenced Packet Exchange (SPX), which give the stack its name, are a direct adaptation of the Xerox XNS protocols, which are much less widespread than the IPX/SPX stack. The popularity of the IPX/SPX stack is directly related to the Novell NetWare operating system, which still retains the world leadership in number installed systems, although recently its popularity has decreased somewhat and its growth rate lags behind Microsoft Windows NT.

Many features of the IPX/SPX stack are due to the orientation of early versions of the NetWare OS (up to version 4.0) for working in small local networks consisting of personal computers with modest resources. It is clear that for such computers Novell needed protocols, the implementation of which would require minimal amount RAM (limited to 640 KB in IBM-compatible computers running MS-DOS) and which would run quickly on processors of low processing power. As a result, the IPX/SPX stack protocols until recently worked well in local networks and not so well in large corporate networks, since they overloaded slow global links with broadcast packets that are intensively used by several protocols in this stack (for example, to establish communication between clients and servers). This circumstance, as well as the fact that the IPX/SPX stack is proprietary to Novell and requires a license to implement it (that is, open specifications were not supported), for a long time limited its distribution only to NetWare networks. However, since the release of NetWare 4.0, Novell has made and continues to make major changes to its protocols aimed at adapting them to work in corporate networks. Now the IPX/SPX stack is implemented not only in NetWare, but also in several other popular network operating systems, for example SCO UNIX, Sun Solaris, Microsoft Windows NT.

NetBIOS/SMB stack

This stack is widely used in products from IBM and Microsoft. All the most common protocols Ethernet, Token Ring, FDDI and others are used at the physical and data link layers of this stack. The NetBEUI and SMB protocols operate at the upper levels.

The NetBIOS (Network Basic Input/Output System) protocol appeared in 1984 as a network extension of the standard functions of the IBM PC Basic Input/Output System (BIOS) for network program PC Network from IBM. This protocol was later replaced by the so-called NetBEUI - NetBIOS Extended User Interface protocol. To ensure application compatibility, the NetBIOS interface was retained as an interface to the NetBEUI protocol. The NetBEUI protocol was designed to be an efficient, low-resource protocol for networks of no more than 200 workstations. This protocol contains many useful network functions, which can be attributed to the network, transport and session layers of the OSI model, but it cannot be used to route packets. This limits the use of the NetBEUI protocol to local networks that are not divided into subnets, and makes it impossible to use it in composite networks. Some of the limitations of NetBEUI are addressed by the NBF (NetBEUI Frame) implementation of this protocol, which is included in the Microsoft Windows NT operating system.

The SMB (Server Message Block) protocol performs the functions of the session, representative and application layers. SMB is used to implement file services, as well as printing and messaging services between applications.

IBM's SNA, Digital Equipment's DECnet, and Apple's AppleTalk/AFP protocol stacks are used primarily in operating systems ah and network equipment of these companies.

In Fig. Figure 1.30 shows the correspondence of some of the most popular protocols to the levels of the OSI model. Often this correspondence is very conditional, since the OSI model is only a guide to action, and quite general, and specific protocols were developed to solve specific problems, and many of them appeared before the development of the OSI model. In most cases, stack developers have prioritized networking speed over modularity—no stack other than the OSI stack is split into seven layers. Most often, 3-4 levels are clearly distinguished in the stack: level network adapters, which implements the protocols of the physical and data link layers, the network layer, the transport layer and the service layer, which incorporates the functions of the session, representative and application layers.

Rice. 1.30. Compliance of popular protocol stacks with the OSI model

	List of abbreviations
1.	Introduction
2.	OSI protocol stack
2.1.	Understanding the OSI Model and the OSI Stack
2.2.	Physical layer
2.3.	Data Link Layer
2.4.	Network layer
2.4.1.	Connectionless services (CLNP/CLNS)
2.4.2.	Connection-based services (CONS/CMNP)
2.4.3.	Addressing
2.5.	Transport layer
2.6.	Session layer
2.7.	Representative level
2.8.	Application layer
3.	Conclusion
4.	List of sources used

List of abbreviations

ACSE (Association Control Service Element) – association control service element

AFI (authority and format identifier) ​​– format and authority identifier

ASE (Application Service Elements) – application level service elements

CASEs (Common-Application Service Elements) – general services application level

CLNP (Connectionless Network Protocol) – connectionless protocol

CLNS (Connectionless Network Service) - connectionless services

CMIP (Common Management Information Protocol) – common management information protocol

CMNP (Connection-Mode Network Protocol) – connection-based protocol

CONS (Connection-Oriented Network Service) – a service that establishes a logical connection

DS ( Directory Services) – directory services

DSP (domain specific part) – domain specific part

ID (Identifier) ​​– identifier

IDI (initial domain identifier) ​​– identifier of the original domain

IDP (initial domain part) – initial part of the domain

IONL (Internal organization of the network level) – internal organization of network levels

ISO (International Standardization Organization) - International Standards Organization

FTAM ( File Transfer Access, and Management) – transfer, access and management of files

GOSIP (Government Open Systems Interconnection Profile) - government standard for open systems interconnection

MHS (Message Handling Systems) – message processing systems

NET (network entity title) – network address

NSAP (network service access point) – access point to network services

OSI (Open Systems Interconnections) - Open Systems Interconnection

PDU (protocol data unit) – protocol information unit

PSAP (Presentation Access Point) - presentation access point

ROSE (Remote Operations Service Element) – service element for gaining access to the operation of a remote device

RTSE (Reliable Transfer Service Element) – reliable transmission service element

SASEs (Specific-Application service elements) – special application level services

SSAP (Session-Service Access Point) – session service access point

VTP (Virtual Terminal Protocol) – virtual terminal protocol

Introduction

Organization of interaction between devices on the network is complex problem, it includes many aspects, starting from the coordination of electrical signal levels, framing, checking checksums and ending with application authentication issues.

In the early years of computer-to-computer communication, networking software was created haphazardly, for each individual case. After networks gained sufficient popularity, some of the developers recognized the need to standardize related software products and hardware development. It was believed that standardization would allow vendors to develop hardware and software systems that could communicate with each other even if they were based on different architectures. With this goal in mind, the International Standards Organization (ISO) began developing the OSI reference model. The OSI reference model was completed and released in 1984.

OSI was a new attempt to create networking standards to ensure interoperability between solutions from different vendors. At that time, many large networks were forced to support multiple communication protocols and included a large number of devices that were unable to communicate with other devices due to the lack of common protocols.

The OSI reference model was a big step in the creation of modern networking concepts. She popularized the idea of ​​a common model of protocols located at different layers that define the interaction between network devices and software.

A distinction must be made between the OSI protocol stack and the OSI model. While the OSI model conceptually defines the procedure for the interaction of open systems, dividing the task into 7 layers, standardizing the purpose of each layer and introducing standard names for the layers, the OSI stack is a set of very specific protocol specifications that form a consistent protocol stack. This protocol stack is supported by the US government in its GOSIP program. All government computer networks installed after 1990 must either directly support the OSI stack or provide a means to migrate to the stack in the future. However, the OSI stack is more popular in Europe than in the US, as Europe has fewer legacy networks installed that use their own protocols. There is also a big need for a common stack in Europe, as there are so many different countries.

This is an international, manufacturer-independent standard. It can enable collaboration between corporations, partners and suppliers. This interaction is complicated by addressing, naming, and data security issues. All these problems are partially solved in the OSI stack.

OSI protocol stack

Understanding the OSI Model and the OSI Protocol Stack

The OSI model has seven layers. The emergence of just such a structure was due to the following considerations.

· A layer should be created as a separate abstraction layer is needed.

· Each level must perform a strictly defined function.

· The selection of functions for each level should be made taking into account the creation of standardized international protocols.

· Boundaries between layers should be chosen so that data flow between interfaces is minimal.

· The number of levels should be high enough so that different functions are not unnecessarily combined into one level, but not too high so that the architecture becomes unwieldy.

For each level, a set of query functions is defined, with which modules at a given level can be accessed by modules at a higher level to solve their problems. This formally defined set of functions performed by a given layer for the layer above it, as well as the message formats exchanged between two neighboring layers during their interaction, is called an interface.

An interface defines the overall service provided by a given layer to the layer above it.

When organizing the interaction of computers on a network, each level conducts “negotiations” with the corresponding level of another computer. When transmitting messages, both participants in a network exchange must accept many agreements. For example, they must agree on the levels and shape of electrical signals, how to determine the length of messages, agree on methods for verifying reliability, etc. In other words, conventions must be adopted at all levels, from the lowest bit transfer level to the highest high level, detailing how the information should be interpreted.

The rules for interaction between two machines can be described as a set of procedures for each level. Such formalized rules that determine the sequence and format of messages exchanged between network components located at the same level, but in different nodes, are called protocols.

From the above definitions, you can see that the concepts “interface” and “protocol” essentially mean the same thing, namely, formalized specified procedures for the interaction of components that solve the problem of connecting computers on a network. However, quite often there is some nuance in the use of these terms: the concept of “protocol” is more often used when describing the rules of interaction between components of the same level located on different nodes of the network, and “interface” - when describing the rules of interaction of components of neighboring levels located within the same node ( Fig. 1 - Interaction between network nodes).

Fig.1. Interaction between network nodes

The OSI protocol suite consists of numerous standard protocols based on the OSI model (Fig. 2 OSI protocols of all levels of the OSI model).

Any protocol of the OSI model must interact either with protocols at its layer, or with protocols one unit higher and/or lower than its layer. Any protocol of the OSI model can perform only the functions of its layer and cannot perform functions of another layer, which is not performed in the protocols of alternative models.

Fig.2. OSI protocols of all levels of the OSI model

©2015-2019 site
All rights belong to their authors. This site does not claim authorship, but provides free use.
Page creation date: 2016-08-20

All of these stacks, except SNA at the lower levels - physical and data link - use the same well standardized protocols Ethernet, Token Ring, FDDI and a number of others, which allow you to use the same equipment in all networks. But at the upper levels, all stacks operate according to their own protocols. These protocols often do not conform to the layering recommended by the OSI model. In particular, the functions of the session and presentation layers are typically combined with the application layer. This discrepancy is due to the fact that the OSI model appeared as a result of a generalization of already existing and actually used stacks, and not vice versa.

OSI stack

A clear distinction should be made between the OSI model and the OSI stack. If the OSI model is a conceptual framework for interconnecting open systems, the OSI stack is a set of very specific protocol specifications.

Unlike other protocol stacks, the OSI stack fully follows the OSI model, it includes protocol specifications for all seven interoperability layers defined in that model. At the lower layers, the OSI stack supports Ethernet, Token Ring, FDDI, WAN protocols, X.25 and ISDN - that is, it uses lower-layer protocols developed outside the stack, like all other stacks. The protocols of the network, transport and session layers of the OSI stack are specified and implemented by various manufacturers, but are not yet widespread. The most popular protocols in the OSI stack are application protocols. These include: FTAM file transfer protocol, VTP terminal emulation protocol, X.500 help desk protocols, X.400 email protocols, and a number of others.

The protocols in the OSI stack are characterized by complexity and ambiguous specifications. These properties were the result of the general policy of the stack developers, who sought to take into account all cases and all existing technologies in their protocols. To this we must also add the consequences of a large number of political compromises that are inevitable when adopting international standards on such a pressing issue as the construction of open computer networks.

Because of their complexity, OSI protocols require a lot of CPU processing power, making them more suitable for powerful machines rather than personal computer networks.

Stack OSI- international standard independent of manufacturers. It is supported by the US government through its GOSIP program, which requires all computer networks installed in US government agencies after 1990 to either directly support the OSI stack, or provide a means to migrate to that stack in the future. However, the OSI stack is more popular in Europe than in the US, as there are fewer legacy networks left in Europe running native protocols. Most organizations are just planning the transition to the OSI stack, and very few have started creating pilot projects. Among those working in this direction are the US Navy Department and the NFSNET network. One of the largest manufacturers supporting OSI is AT&T, its Stargroup network is entirely based on this stack.

TCP/IP stack

The TCP/IP stack was developed at the initiative of the US Department of Defense more than 20 years ago to connect the experimental ARPAnet network with other networks as a set of common protocols for heterogeneous computing environments. A great contribution to the development of the TCP/IP stack, which got its name from the popular IP and TCP protocols, was made by specialists from Berkeley University, who implemented the stack protocols in the UNIX OS version. The popularity of this operating system led to the widespread adoption of TCP, IP and other protocol stacks. Today, this stack is used to connect computers on the Internet, as well as in a huge number of corporate networks.

The TCP/IP stack at the lower level supports all popular standards of the physical and data link layers: for local networks - Ethernet, Token Ring, FDDI, for global networks - protocols for working on analog dial-up and leased lines SLIP, PPP, protocols for territorial networks X.25 and ISDN.

The main protocols of the stack, which give it its name, are IP and TCP. These protocols, in OSI model terminology, belong to the network and transport layers, respectively. IP ensures that the packet travels across the composite network, and TCP ensures the reliability of its delivery.

Over many years of use in networks of various countries and organizations, the TCP/IP stack has incorporated a large number of application-level protocols. These include such popular protocols as the FTP file transfer protocol, the telnet terminal emulation protocol, the SMTP mail protocol used in Internet e-mail, hypertext services of the WWW service and many others.

Today, the TCP/IP stack is one of the most common transport protocol stacks in computer networks.

Indeed, the Internet alone connects about 10 million computers around the world that interact with each other using the TCP/IP protocol stack.

The rapid growth of Internet popularity has also led to changes in the balance of power in the world. communication protocols- TCP/IP protocols, on which the Internet is built, began to quickly crowd out the undisputed leader of past years - Novell's IPX/SPX stack. Today in the world, the total number of computers on which the TCP/IP stack is installed has exceeded the number of computers on which the IPX/SPX stack is running, and this indicates a change in the attitude of local network administrators to the protocols used on desktop computers, since they were previously used Almost everywhere the Novell protocols needed to access NetWare file servers worked. The process of promoting the TCP/IP stack to a leading position in all types of networks continues, and now any industrial operating system must include a software implementation of this stack.

Although the TCP/IP protocols are inextricably linked to the Internet, and each of the multimillion-dollar armada of Internet computers runs on this stack, there are a large number of local, corporate and territorial networks that are not directly part of the Internet that also use the TCP/IP protocols. To distinguish these networks from the Internet, they are called TCP/IP networks or simply IP networks.

Because the TCP/IP stack was originally designed for the global Internet, it has many features that give it an edge over other protocols when it comes to building networks that include wide-area communications. In particular, a very useful feature that makes this protocol useful in large networks is its ability to fragment packets. Indeed, a complex composite network often consists of networks built on completely different principles. Each of these networks can set its own value for the maximum length of a unit of transmitted data (frame). In this case, when moving from one network with a larger maximum length to another with a smaller maximum length, it may be necessary to split the transmitted frame into several parts. The IP protocol of the TCP/IP stack effectively solves this problem.

Another feature of TCP/IP technology is its flexible addressing system, which makes it easier to include networks of other technologies into the internet (internet or composite network) compared to other protocols of similar purposes. This property also facilitates the use of the TCP/IP stack for building large heterogeneous networks.

The TCP/IP stack uses broadcast capabilities very sparingly. This property is simply necessary when working on slow communication channels typical of territorial networks.

However, the price to pay for the advantages here is high resource requirements and the complexity of administering IP networks. To realize powerful functionality TCP/IP stack protocols require large computational costs. A flexible addressing system and the refusal of broadcasts lead to the presence in the IP network of various centralized services such as DNS, DHCP, etc. Each of these services simplifies network administration and equipment configuration, but at the same time, it itself requires close attention from administrators .

You can give other arguments for and against, but the fact remains that today TCP/IP is the most popular protocol stack, widely used in both global and local networks.

IPX/SPX stack

This stack is the original Novell protocol stack, developed for the NetWare network operating system back in the early 80s. The network and session layer protocols Internetwork Packet Exchange (IPX and Sequenced Packet Exchange, SPX), which give the stack its name, are a direct adaptation of Xerox's XNS protocols, which are much less common than the IPX/SPX stack.

The popularity of the IPX/SPX stack is directly related to the Novell NetWare operating system, which for a long time maintained the world leadership in the number of installed systems, although recently its popularity has decreased significantly, and its growth rate is noticeably behind Microsoft Windows NT.

Many features of the IPX/SPX stack are due to the orientation of early versions of the NetWare OS (up to version 4.0) for working in small local networks consisting of personal computers with modest resources. It is clear that for such computers, Novell needed protocols that would require a minimum amount of RAM (limited in IBM-compatible computers running MS-DOS with a capacity of 640 KB) and that would run quickly on processors of low processing power. As a result, the IPX/SPX stack protocols until recently worked well in local networks and not so well in large corporate networks, since they overloaded slow global links with broadcast packets that are intensively used by several protocols in this stack (for example, to establish communication between clients and servers). This circumstance, as well as the fact that the IPX/SPX stack is proprietary to Novell and must be licensed to implement it (that is, open specifications were not supported), for a long time limited its scope of activity only to networks

Protocol stacks

A protocol stack is a hierarchically organized set of network protocols at various levels, sufficient to organize and ensure the interaction of nodes in the network. Currently, networks use a large number of communication protocol stacks. The most popular stacks are: TCP/IP, IPX/SPX, NetBIOS/SMB, Novell NetWare, DECnet, XNS, SNA and OSI. All of these stacks, except for SNA, at the lower levels - physical and data link - use the same well-standardized protocols Ethemet, Token Ring, FDDI and some others, which allow the same equipment to be used in all networks. But at the upper levels, all stacks operate according to their own protocols. These protocols often do not conform to the layering recommended by the OSI model. In particular, the functions of the session and presentation layers are typically combined with the application layer. This discrepancy is due to the fact that the OSI model appeared as a result of a generalization of already existing and actually used stacks, and not vice versa.

All protocols included in the stack were developed by one manufacturer, that is, they are able to work as quickly and efficiently as possible.

An important point in the operation of network equipment, in particular a network adapter, is the binding of protocols. It allows you to use different protocol stacks when servicing one network adapter. For example, you can use TCP/IP and IPX/SPX stacks simultaneously. If suddenly an error occurs when trying to establish a connection with the recipient using the first stack, then a switch to using the protocol from the next stack will automatically occur. An important point in in this case is the binding order, since it clearly affects the use of one or another protocol from different stacks.

Regardless of how many network adapters are installed in the computer, binding can be carried out either “one to several” or “several to one”, that is, one protocol stack can be tied to several adapters at once or several stacks to one adapter.

NetWare is a network operating system and a set of network protocols that are used in this system to interact with client computers connected to the network. The system's network protocols are based on the XNS protocol stack. NetWare currently supports TCP/IP and IPX/SPX protocols. Novell NetWare was popular in the 80s and 90s due to its greater efficiency compared to operating systems general purpose. This is now an outdated technology.

The XNS (Xerox Network Services Internet Transport Protocol) protocol stack was developed by Xerox for transmitting data over Ethernet networks. Contains 5 levels.

Level 1 - transmission medium - implements the functions of the physical and data link layers in the OSI model:

* manages data exchange between the device and the network;

* routes data between devices on the same network.

Layer 2 - internetwork - corresponds to the network layer in the OSI model:

* manages data exchange between devices located on different networks (provides datagram service in terms of the IEEE model);

* describes the way data flows through the network.

Layer 3 - transport - corresponds to the transport layer in the OSI model:

* provides end-to-end communication between the data source and destination.

Level 4 - control - corresponds to the session and representative levels in the OSI model:

* controls the presentation of data;

* manages control over device resources.

Level 5 - application - corresponds to the highest levels in the OSI model:

* provides data processing functions for application tasks.

The TCP/IP (Transmission Control Protocol/Internet Protocol) protocol stack is the most common and functional today. It works in local networks of any size. This stack is the main stack on the global Internet. Stack support was implemented in computers with an operating system UNIX system. As a result, the popularity of the TCP/IP protocol has increased. The TCP/IP protocol stack includes quite a lot of protocols operating at different levels, but it got its name thanks to two protocols - TCP and IP.

TCP (Transmission Control Protocol) is a transport protocol designed to control data transmission in networks using the TCP/IP protocol stack. IP (Internet Protocol) is a network layer protocol designed to deliver data over a composite network using one of the transport protocols, such as TCP or UDP.

The lower layer of the TCP/IP stack uses standard protocols data transfer, which makes it possible to use it in networks using any network technologies and on computers with any operating system.

The TCP/IP protocol was originally developed for use in global networks, which is why it is extremely flexible. In particular, thanks to the ability to fragment packets, data, despite the quality of the communication channel, in any case reaches the addressee. In addition, thanks to the presence of the IP protocol, data transfer between dissimilar network segments becomes possible.

The disadvantage of the TCP/IP protocol is the complexity of network administration. Thus, for the normal functioning of the network, additional servers are required, such as DNS, DHCP, etc., maintaining the operation of which takes up most of the system administrator’s time. Limoncelli T., Hogan K., Cheylap S. - System and network administration. 2nd ed. year 2009. 944с

The IPX/SPX (Internetwork Packet Exchange/Sequenced Packet Exchange) protocol stack is developed and owned by Novell. It was developed for the needs of the Novell NetWare operating system, which until recently occupied one of the leading positions among server operating systems.

The IPX and SPX protocols operate at the network and transport layers of the ISO/OSI model, respectively, and therefore complement each other perfectly.

The IPX protocol can transmit data using datagrams using network routing information. However, in order to transmit data along the found route, a connection must first be established between the sender and the recipient. This is what the SPX protocol or any other transport protocol that works in tandem with IPX does.

Unfortunately, the IPX/SPX protocol stack is initially designed to serve small networks, so its use in large networks is ineffective: excessive use of broadcasting on low-speed communication lines is unacceptable.

At the physical and data link layers, the OSI stack supports the Ethernet, Token Ring, FDDI protocols, as well as the LLC, X.25 and ISDN protocols, that is, it uses all the popular lower-layer protocols developed outside the stack, like most other stacks. The network layer includes the relatively rarely used Connectionoriented Network Protocol (CONP) and Connectionless Network Protocol (CLNP). The routing protocols of the OSI stack are ES-IS (End System -- Intermediate System) between end and intermediate systems and IS-IS (Intermediate System -- Intermediate System) between intermediate systems. The transport layer of the OSI stack hides the differences between connection-oriented and connectionless network services so that users receive the desired quality of service regardless of the underlying network layer. To provide this, the transport layer requires the user to specify the desired quality of service. Application layer services provide file transfer, terminal emulation, directory services, and mail. Of these, the most popular are directory services (X.500 standard), Email(X.400), Virtual Terminal Protocol (VTP), File Transfer, Access and Management Protocol (FTAM), Forwarding and Job Management Protocol (JTM).

A fairly popular protocol stack developed by IBM and Microsoft, respectively, aimed at use in the products of these companies. Like TCP/IP, standard protocols such as Ethernet, Token Ring and others operate at the physical and data link levels of the NetBIOS/SMB stack, which makes it possible to use it in conjunction with any active network equipment. At the upper levels, the NetBIOS (Network Basic Input/Output System) and SMB (Server Message Block) protocols operate.

The NetBIOS protocol was developed in the mid-80s of the last century, but was soon replaced by the more functional NetBEUI (NetBIOS Extended User Interface) protocol, which allows for very efficient information exchange in networks consisting of no more than 200 computers.

To exchange data between computers, logical names are used that are assigned to computers dynamically when they are connected to the network. In this case, the name table is distributed to each computer on the network. It also supports working with group names, which allows you to transfer data to several recipients at once.

The main advantages of the NetBEUI protocol are speed and very low resource requirements. If you need to organize fast data exchange in a small network consisting of a single segment, better protocol Can't find one for this. In addition, to deliver messages established connection is not a mandatory requirement: in the event of no connection, the protocol uses the datagram method, where the message is equipped with the address of the recipient and the sender and “goes on the road”, moving from one computer to another.

However, NetBEUI also has significant drawback: It is completely devoid of any concept of packet routing, so its use in complex composite networks does not make sense. Pyatibratov A.P., Gudyno L.P., Kirichenko A.A. Computers, networks and telecommunication systems Moscow 2009. 292s

As for the SMB (Server Message Block) protocol, it is used to organize network operation at the three highest levels - session, presentation and application levels. It is when you use it that access to files, printers and other network resources becomes possible. This protocol has been improved several times (three versions have been released), which makes it possible to use it even in modern operating systems such as Microsoft Vista and Windows 7. The SMB protocol is universal and can work in tandem with almost any transport protocol, such as TCP/IP and SPX.

The DECnet (Digital Equipment Corporation net) protocol stack contains 7 layers. Despite the difference in terminology, the DECnet layers are very similar to the OSI model layers. DECnet implements the DNA (Digital Network Architecture) concept of network architecture, developed by DEC, according to which heterogeneous computing systems (computers of different classes), operating under different operating systems, can be combined into geographically distributed information and computing networks.

IBM's SNA (System Network Architecture) protocol is designed for remote communication with large computers and contains 7 levels. SNA is based on the host machine concept and provides remote terminal access to IBM mainframes. The main distinguishing feature of SNA is the ability of each terminal to access any application program of the host computer. System network architecture implemented on the basis of a virtual telecommunication access method (VTAM) in the host computer. VTAM manages all communications links and terminals, with each terminal having access to all application programs.

Information exchange is a multifunctional process. Related functions are grouped by purpose and these groups are called "levels of interaction." Unification of levels allows you to create heterogeneous networks with complex topologies. The unification is based on the concept of a reference network model. The model as such only describes the order of network interaction, which is implemented in the form of a protocol stack.

Exchange of information between networked computers is a very complex task. This is due to the fact that there are many hardware and software manufacturers computing systems. The only way out is to unify the means of interfacing systems, namely to use open systems. An open system interacts with other systems based on common publicly available standards and specifications.

In 1984 The International Organization for Standardization (ISO) has introduced an industry standard - open systems interaction model(Open System Interconnection Reference Model - OSI/RM, in Soviet literature - EMVOS) to help vendors create compatible network hardware and software. In accordance with this model, the following levels are distinguished (Fig. 1):

Rice. 1. OSI reference model

• physical (Physical);
• channel (Data Link);
• network (Network);
• transport (Transport);
• session (Session);
• representative (Presentation);
• applied (Application).

According to the OSI reference model, these layers interact as shown in Fig. 2. Thus, the complex task of exchanging information between computers on a network is divided into a number of relatively independent and less complex subtasks interactions between adjacent levels.

Rice. 2. Interaction between OSI layers

Communication between the layers of two network nodes ( horizontal interaction) is carried out in accordance with the unified rules - interaction protocols

IN autonomous system transfer of data between levels ( vertical interaction) is implemented through interfaces API

The boundary between the session and transport layers can be thought of as the boundary between application-layer protocols and lower-layer protocols. If the application, presentation and session layers provide the application processes of the interaction session, then the four lower layers solve the problems of data transportation.

The two lowest levels - physical and channel - are implemented by hardware and software, the remaining five higher levels are implemented, as a rule, by software.

When transmitting information from an application process to the network to the physical layer, it is processed, which consists of dividing the transmitted data into separate blocks, transforming the form of representation or encoding of the data in the block and adding to each block header(header) of the appropriate level (see example). Each header characterizes the data processing protocol used, and each layer perceives as data the entire block received from the previous layer, including the attached header. This construction of the reference model allows us to lay down ( encapsulate) in each information block transmitted over the physical medium, information necessary for selecting a sequence of protocols for implementing reverse transformations on the receiving side.

Physical layer

This layer defines the mechanical, electrical, procedural, and functional characteristics of establishing, maintaining, and releasing physical connections between end systems. The physical layer defines connection characteristics such as voltage levels, timing and physical data rates, maximum transmission distances, connector design parameters, and other similar characteristics. Well-known standards RS-232-C, V.24 and IEEE 802.3 (Ethernet).

Data Link Layer

Data link layer (data link level, information link layer) is responsible for reliable data transmission through a physical channel, namely:

• provides physical addressing (as opposed to network or logical addressing);
• provides error detection in transmission and data recovery;
• monitors the network topology and ensures discipline in the use of the network channel by the end system;
• provides notification of faults;
• provides orderly delivery of data blocks and control of information flow.

For a LAN, the link layer is divided into two sublevels:

• LLC (Logical Link Control) - provides control of a logical link, i.e. the actual functions of the link layer;
• MAC (Media Access Control) - provides special methods for accessing the distribution media.

Network layer

This layer provides connectivity and route selection between two end systems connected to different subnets (segments), which may be separated by multiple subnets and may be located in different geographic locations. Routing protocols allow a network of routers to select optimal routes across interconnected subnets.

Transport layer

The transport layer provides data transport services to higher layers, namely:

• ensures reliable data transportation through the interconnected network;
• provides mechanisms for establishing, maintaining, and orderly termination of virtual channels;
• provides detection and elimination of transportation faults;
• ensures that the end system is not overloaded with too much data.

In other words, the transport layer provides an interface between processes and the network, establishes logical channels between processes and ensures the transmission of information blocks over these channels. These logical channels are called transport channels.

Session layer

The session layer implements the establishment, maintenance and termination of a session of interaction between subscriber application processes. The session layer synchronizes the dialogue between objects of the representative layer, defines synchronization points for intermediate control and recovery during file transfers. This level also allows data exchange in the mode specified application program, or provides the ability to select the exchange mode.

In addition to the basic dialog control function, the session layer provides facilities for class of service selection and exception notification (session, presentation, and application layer problems).

Representative level

The representative level (data presentation level) defines the syntax, formats and structures for presenting the transmitted data (but does not affect the semantics, the meaning of the data). In order for information sent from the application layer of one system to be readable at the application layer of another system, the representative layer translates between known information presentation formats by using a unified information presentation format.

Thus, this layer provides service operations, selected at the application layer, to interpret the data transmitted and received: communication control, data display and structured data management. This service data allows different types of terminals and computing devices to be linked together. An example of a protocol at this layer is XDR.

Application layer

Unlike other layers, the application layer—the OSI layer closest to the user—does not provide services to other OSI layers, but it does provide application processes that are outside the scope of the OSI model.

The application layer provides direct support for application processes and end-user programs (DBMS, word processors, bank terminal programs, etc.) and managing the interaction of these programs with the data network:

• identifies and establishes the presence of prospective communication partners;
• synchronizes jointly working application programs;
• Establishes agreement on procedures for error resolution and information integrity management;
• determines the sufficiency of available resources for the proposed connection.

The OSI model is not an implementation; it only suggests an order for organizing interactions between system components. The implementations of these rules are protocol stacks.

Protocol stacks

OSI stack

The protocols of the OSI stack and their distribution among the levels of the network model are shown in Fig. 3.

NetBIOS/SMB stack

Microsoft and IBM worked together on networking tools for personal computers, so the NetBIOS/SMB protocol stack is their joint brainchild. NetBIOS tools appeared in 1984 as a network extension of the standard functions of the basic input/output system (BIOS) of the IBM PC for the IBM PC Network network program, which at the application level (Fig. 4) used the SMB protocol to implement network services.

Protocol NetBIOS works at three levels of the open systems interaction model: network, transport and session. NetBIOS can provide a higher level of service than the IPX and SPX protocols, but does not have routing capabilities. Thus, NetBIOS is not a network protocol in the strict sense of the word. NetBIOS contains many useful networking functions that can be attributed to the network, transport and session layers, but it cannot be used to route packets, since the NetBIOS frame exchange protocol does not introduce such a concept as a network. This limits the use of the NetBIOS protocol to local networks that are not subnetted. NetBIOS supports both datagram and connection-based communications.

Protocol SMB, corresponding to the application and representative levels of the OSI model, regulates the interaction of the workstation with the server. SMB functions include the following operations:

• Session management. Creation and breaking of a logical channel between the workstation and the network resources of the file server.
• File access. Work station can contact the file server with requests to create and delete directories, create, open and close files, read and write to files, rename and delete files, search for files, retrieve and install file attributes, blocking records.
• Printing service. The workstation can queue files for printing on the server and obtain information about the print queue.
• Messaging service. SMB supports simple messaging with the following functions: send a simple message; send a broadcast message; send start of message block; send message block text; send end of message block; forward username; cancel the shipment; get the machine name.

Because of the large number of applications that use the API functions provided by NetBIOS, many network operating systems implement these functions as an interface to their transport protocols. NetWare has a program that emulates NetBIOS functions based on the IPX protocol, and there are software emulators for NetBIOS for Windows NT and the TCP/IP stack.

TCP/IP stack

The TCP/IP stack, also called the DoD stack and the Internet stack, is one of the most popular communication protocol stacks. The stack was developed at the initiative of the US Department of Defense (DoD) to connect the experimental ARPAnet network with other satellite networks as a set of common protocols for a heterogeneous computing environment. The ARPA network supported developers and researchers in military fields. In the ARPA network, communication between two computers was carried out using the Internet Protocol (IP), which is still the main protocol in the TCP / IP stack and appears in the name of the stack.

Berkeley University made a major contribution to the development of the TCP/IP stack by implementing stack protocols in its version of the UNIX OS. The widespread adoption of the UNIX operating system also led to the widespread adoption of IP and other stack protocols. Worldwide works on the same stack information network Internet, whose division, the Internet Engineering Task Force (IETF), is a major contributor to the improvement of stack standards published in the form of RFC specifications.

Since the TCP/IP stack was developed before the advent of the ISO/OSI open systems interconnection model, although it also has a multi-level structure, the correspondence of the TCP/IP stack levels to the levels of the OSI model is rather conditional.

The structure of the TCP/IP protocols is shown in Fig. 5. TCP/IP protocols are divided into 4 levels.

The lowest (level IV) - the level of gateway interfaces - corresponds to the physical and data link layers of the OSI model. This level in the TCP/IP protocols is not regulated, but supports all popular standards of the physical and data link layer: for local channels these are Ethernet, Token Ring, FDDI, for global channels - their own protocols for operating on analog dial-up and leased lines SLIP/PPP, which establish point-to-point connections via WAN serial links, and X.25 and ISDN WAN protocols. A special specification has also been developed defining the use ATM technologies as a link layer transport.

The next layer (layer III) is the internetworking layer, which deals with the transmission of datagrams using various local area networks, X.25 area networks, ad hoc links, etc. As the main network layer protocol (in terms of the OSI model) in the stack protocol used IP, which was originally designed as a protocol for transmitting packets in composite networks consisting of a large number of local networks, united by both local and global connections. Therefore, the IP protocol works well in networks with complex topologies, rationally using the presence of subsystems in them and economically spending throughput low-speed communication lines. The IP protocol is a datagram protocol.

The level of internetworking also includes all protocols related to the compilation and modification of routing tables, such as protocols for collecting routing information R.I.P.(Routing Internet Protocol) and OSPF(Open Shortest Path First), as well as the Internet Control Message Protocol ICMP(Internet Control Message Protocol). The latter protocol is designed to exchange error information between the router and the gateway, the source system and the destination system, that is, to organize feedback. Using special ICMP packets, it is reported that it is impossible to deliver a packet, that the lifetime or duration of assembling a packet from fragments has been exceeded, anomalous parameter values, a change in the forwarding route and type of service, the state of the system, etc.

The next level (level II) is called basic. The transmission control protocol operates at this level TCP(Transmission Control Protocol) and User Datagram Protocol UDP(User Datagram Protocol). The TCP protocol provides a stable virtual connection between remote application processes. The UDP protocol ensures the transmission of application packets using the datagram method, that is, without establishing a virtual connection, and therefore requires less overhead than TCP.

The top level (level I) is called application. Over many years of use in the networks of various countries and organizations, the TCP/IP stack has accumulated a large number of protocols and application level services: FTP file copy protocol, telnet and ssh remote control protocols, SMTP mail protocol, hypertext services for accessing remote information, such as WWW and a lot others. Let's briefly look at some of the stack protocols that are most closely related to the topics of this course.

Protocol SNMP(Simple Network Management Protocol) is used to organize network management. The management problem is divided here into two problems. The first task is related to the transfer of information. Control information transfer protocols determine the procedure for interaction between the server and the client program running on the administrator's host. They define the message formats that are exchanged between clients and servers, as well as the formats for names and addresses. The second challenge is related to controlled data. The standards regulate what data should be stored and accumulated in gateways, the names of this data, and the syntax of these names. The SNMP standard defines a specification information base network management data. This specification, known as the Management Information Base (MIB), defines the data elements that a host or gateway must store and the permissible operations on them.

File Transfer Protocol FTP(File Transfer Protocol) implements remote access to the file. To ensure reliable transmission, FTP uses the connection-oriented protocol TCP as its transport. In addition to file transfer protocol, FTP offers other services. This gives the user the opportunity interactive work with a remote machine, for example, it can print the contents of its directories; FTP allows the user to specify the type and format of the data to be stored. Finally, FTP authenticates users. Before accessing the file, protocol requires users to provide their username and password.

In the TCP/IP stack FTP protocol offers the widest range of services for working with files, but it is also the most difficult to program. Applications that do not require the full capabilities of FTP can use another, more cost-effective protocol, the Simple File Transfer Protocol. TFTP(Trivial File Transfer Protocol). This protocol only implements file transfer, and the transport used is a simpler than TCP, connectionless protocol - UDP.

Protocol telnet provides the transfer of a stream of bytes between processes, as well as between a process and a terminal. Most often, this protocol is used to emulate a remote computer terminal.

Control questions

1. What is the OSI model for?
2. List the layers of the OSI model
3. What problems does the application layer of the OSI model solve?
4. What problems does the presentation layer of the OSI model solve?
5. What tasks does the transport layer of the OSI model solve?
6. What problems does the network layer of the OSI model solve?
7. What problems does the data link layer of the OSI model solve?
8. What problems does the physical layer of the OSI model solve?
9. How does the OSI model exchange data between layers?
10. What is a "protocol stack"

Permanent address of this page: