Basic technology of local computing networks. Network technologies

Local networking architectures or technology can be divided into two generations. The first generation includes architectures that provide low and average information transfer rate: Ethernet 10 Mbps), Token Ring (16 Mbps) and Arc Net (2.5 Mbps).

To transfer data, these technologies use cables with copper residential. To the second generation of technologies includes modern high-speed architectures: FDDI (100 Mbit / s), atm (155 Mbps) and upgraded versions of the first generation architectures (Ethernet): Fast Ethernet (100 Mbps) and Gigabit Ethernet (1000 Mbps ). Advanced options for first generation architectures are designed to use cables with copper cores and on fiber-optic data lines. New technologies (FDDI and ATM) are focused on using fiber-optic data transmission lines and can be used to simultaneously transmit information of various types (video images, voices and data). Network technology is a minimum set of standard protocols and implementing their software and hardware, sufficient to build a computing network. Network technologies are called basic technologies. Currently, there are a huge number of networks that have different levels of standardization, such well-known technologies such as Ethernet, token-ring, ArcNet, FDDI received widespread.

Network access methods

Ethernet It is the method of multiple access with listening to carrier and resolution of collisions (conflicts). Before the start of the transfer, each workstation determines whether the channel is free or busy. If the channel is free, the station begins to transmit data. Really conflicts lead to a reduction in network speed only in the case when 80-100 stations are operating. Access method Arcnet. This access method was widespread mainly due to the fact that ArcNet equipment is cheaper than Ethernet or token -ring equipment. Arcnet is used in local networks with a star topology. One of the computers creates a special marker (special message), which is consistently transmitted from one computer to another. If the station must transfer the message, it, having received a marker, generates a package, complemented by the addresses of the sender and destination. When the package comes to the destination station, the message is "uncovered" from the marker and passes the station. Access method Token Ring.. This method is developed by IBM; It is calculated by the network topology network. This method resembles ArcNet, as it also uses a marker transmitted from one station to another. Unlike ArcNet when accessing Token Ring, it is possible to assign different priorities to different workstations.

Basic technology LAN

Ethernet technology is now most popular in the world. The classic Ethernet network uses a standard coaxial cable of two types (thick and thin). However, the Ethernet version of the ethernet is becoming increasingly used as a medium of twisted pairs, since the installation and maintenance of them is much easier. Topologies type "tire" and type "passive star" are used. The standard defines four main types of transmission media.

 10Base5 (thick coaxial cable);

 10Base2 (thin coaxial cable);

 10Base-T (twisted pair);

 10Base-F (fiber optic cable).

Fast Ethernet - high-speed variation of the Ethernet network, providing a transmission rate of 100 Mbps. Fast Ethernet Network Compatible with Networks made according to Ethernet. The main topology of the Fast Ethernet network is a passive star.

The standard defines three types of transmission media for FAST Ethernet:

 100Base-T4 (quad twisted pair);

 100Base-TX (dual twisted pair);

 100Base-FX (fiber optic cable).

Gigabit Ethernet - high-speed variation of the Ethernet network, providing 1000 Mbps transmission speed. The Gigabit Ethernet network standard currently includes the following types of transmission media:

 1000Base-SX is a segment on a multimode fiber optic cable with a light wavelength of 850 nm.

 1000Base-LX is a segment on a multimode and single-mode fiber optic cable with a wavelength of a luminous signal of 1300 nm.

 1000Base-CX is a segment on an electrical cable (shielded twisted pair).

 1000Base-T - segment on the electrical cable (quad-stable unshielded twisted pair).

Due to the fact that the networks are compatible, easily and simply connect the Ethernet segments, Fast Ethernet and Gigabit Ethernet into a single network.

The Taken-Ring network is offered by IBM. Token-Ring was intended to combine all types of computers manufactured by IBM (from personal to large). Taken-Ring network has a star-ring topology. The ArcNet network is one of the oldest networks. As a topology, the ARCNET network uses the "tire" and "passive star". The ArcNet network has enjoyed great popularity. Among the main advantages of the ARCnet network, you can call high reliability, low cost of adapters and flexibility. The main disadvantages of the network is the low speed of information transfer (2.5 Mbps). FDDI (Fiber Distributed Data Interface) -standardized specification for high-speed data network architecture on fiber-optic lines. Transmission rate - 100 Mbps. The main technical characteristics of the FDDI network are as follows:

 The maximum number of network subscribers is 1000.

 Maximum length of the network ring - 20 km

 The maximum distance between network subscribers is 2 km.

 Transmission environment - fiber optic cable

 Mode of access - marker.

 Information transfer rate is 100 Mbps.

Introduction ........................................................................ ..3

1 network Ethernet and Fast Ethernet .................................... 5

2 network token-ring ............................................................ .9

3 SET ARCNET .................................................................. .14

4 network FDDI ...................................................................................... 18

5 network 100VG-AnyLAN ...........................................................................................................................................

6 ultra-speed networks ................................................ .25

7 Wireless Networks ...................................................... .31

Conclusion ........................................................................36

List of sources used ........................... 39

Introduction

During the time that has passed since the appearance of the first local networks, several hundred cells have been developed for a wide variety of network technologies, but few have received noticeable distribution. This is primarily due to the high level of standardization of the principles of the organization of networks and with their support-known companies. However, the standard networks do not always have record characteristics, provide the most optimal exchange modes. But large volumes of release of their equipment and, therefore, its low cost give them huge benefits. It is also important that software manufacturers are also primarily oriented on the most common networks. Therefore, the user chooses the standard networks has a complete guarantee of the compatibility of equipment and programs.

The purpose of this course work is to consider the existing technologies of local networks of their characteristics and advantages or deficiencies in front of each other.

I chose the technology of local networks, because in my opinion, this topic is now particularly relevant when mobility, speed and convenience are appreciated worldwide, with the smallest time as possible.

Currently, a decrease in the number of types of networks used has become a trend. The fact is that an increase in the transfer rate in local networks up to 100 and even up to 1000 Mbps requires the use of the most advanced technologies, conducting expensive scientific research. Naturally, it can afford only the largest firms that support their standard networks and their more advanced varieties. In addition, a large number of consumers have already established some networks and does not want to completely replace network equipment at once. In the near future, it is hardly worth expecting that fundamentally new standards will be adopted.

The market offers standard local networks of all possible topologies, so the choice of users has. Standard networks provide a wide range of permissible network size, number of subscribers and, not less important, prices for equipment. But the choice is still not easy. After all, in contrast to software, replaced which is not difficult, the equipment usually serves for many years, its replacement leads not only to considerable costs, to the need to drive cables, but also to the revision of the system of computer tools of the organization. In this regard, errors in the choice of equipment are usually costs much more than errors when choosing software.

1 Ethernet and Fast Ethernet

The highest distribution among standard networks received an Ethernet network. For the first time it appeared in 1972 (the developer was the well-known Xerox firm). The network was quite successful, and as a result of this in 1980, such largest companies as DEC and Intel were supported. Their efforts in 1985, the Ethernet network became an international standard, it was adopted by the largest international organizations on standards: IEEE and ELECTERONIC Engineers Committee (ECMA (European Computer Manufacturers Association).

The standard was called IEEE 802.3 (in English is read as "Eight Oh Two Dot Three"). It defines multiple access to a bus type monocanal with conflict detection and transmission control. Some other networks satisfy this standard, since the level of its detail is low. As a result of the IEEE 802.3 standard, both constructive and electrical characteristics were often incompatible. However, recently, the IEEE 802.3 standard is considered to be the standard Ethernet network.

The main characteristics of the initial standard IEEE 802.3:

• topology - tire;
• transmission medium - coaxial cable;
• transmission rate - 10 Mbps;
• maximum network length - 5 km;
• maximum number of subscribers - up to 1024;
• network segment length - up to 500 m;
• number of subscribers on one segment - up to 100;
• access method - CSMA / CD;
• the transmission is narrow-band, that is, without modulation (monocanal).

Strictly speaking, there are minor differences between IEEE 802.3 and Ethernet standards, but they usually prefer not to remember.

Ethernet network is now most popular in the world (more than 90% of the market), it is alleged that it will remain in the coming years. This consistently contributed to the fact that from the very beginning, the characteristics, parameters, network protocols were discovered from the very beginning, as a result of which the huge number of manufacturers around the world began to produce Ethernet equipment, fully compatible with each other.

In the classical Ethernet network, a 50-ohm coaxial cable of two types (thick and thin) was used. However, recently (from the beginning of the 90s), the highest distribution received the Ethernet version using twisted pairs as a medium. The standard is also defined for the application of the fiber optic cable. To account for these changes to the initial standard IEEE 802.3, appropriate additions were made. In 1995, an additional standard appeared on a faster version of Ethernet operating at 100 Mbit / s (the so-called Fast Ethernet, IEEE 802.3u standard), using a twin or fiber-optic cable as a medium. In 1997, the version for the speed of 1000 Mbps (Gigabit Ethernet, IEEE 802.3z standard) appeared.

In addition to the standard topology, the tire is increasingly used topologies such as passive star and passive tree.

Classical Ethernet Topology

The maximum length of the network cable as a whole (the maximum signal path) theoretically can reach 6.5 kilometers, but practically does not exceed 3.5 kilometers.

The FAST Ethernet network does not provide a tire physical topology, only a passive star or passive tree is used. In addition, Fast Ethernet has much more stringent requirements for the maximum length of the network. After all, with an increase in 10 times the transmission rate and preservation of the package format, its minimum length becomes ten times shorter. Thus, 10 times the permissible value of the double time of the signal over the network is reduced (5.12 μs against 51.2 μs in Ethernet).

For information transfer to the Ethernet network uses a standard manchester code.

Access to the Ethernet network is carried out by random CSMA / CD method that ensures subscriber equality. The network uses variable length packets with a structure.

For an Ethernet network operating at a speed of 10 Mbps, the standard defines the four main types of network segments focused on different information transfer environments:

• 10Base5 (thick coaxial cable);
• 10Base2 (thin coaxial cable);
• 10Base-T (twisted pair);
• 10Base-FL (fiber optic cable).

The name of the segment includes three items: the figure "10" means the transmission rate of 10 Mbps, the word Base - transmission in the main frequency band (that is, without modulating a high-frequency signal), and the last element is the permissible length of the segment: "5" - 500 meters, "2" - 200 meters (more precisely, 185 meters) or communication type: "T" - twisted pair (from English "twisted-pair"), "F" - fiber optic cable (from English "Fiber Optic").

In the same way for the Ethernet network operating at a speed of 100 Mbps (Fast Ethernet), the standard defines three types of segments that differ in the type of transmission medium:

• 100Base-T4 (quad twisted pair);
• 100Base-TX (dual twisted pair);
• 100Base-FX (fiber optic cable).

Here the figure "100" means the transfer rate of 100 Mbit / s, the letter "T" is a twisted pair, the letter "F" - the fiber optic cable. Types 100Base-TX and 100Base-FX are sometimes combined under the name 100Base-X, and 100Base-T4 and 100Base-TX - under the name 100Base-T.

The development of Ethernet technology goes along the path of increasingly departing from the initial standard. The use of new transmission and switched media allows you to significantly increase the size of the network. The refusal to manchester code (on the Fast Ethernet and Gigabit Ethernet network) provides an increase in the data transfer rate and reduce the requirements for the cable. Refusal from the CSMA / CD control method (with full-duplex exchange mode) makes it possible to dramatically improve the efficiency of work and remove restrictions from the network length. However, all new network varieties are also called an Ethernet network.

2 network token-ring

The Taken-Ring network (marker ring) was proposed by IBM in 1985 (the first option appeared in 1980). It was intended to combine all types of computers manufactured by IBM. Already the fact that it supports IBM, the largest manufacturer of computer equipment, indicates that she needs to pay special attention. But no less important is that the token-ring is currently the international standard IEEE 802.5 (although there is minor differences between token-ring and IEEE 802.5). This puts this network for one level by status with Ethernet.

Taken-Ring was developed as a reliable Ethernet alternative. And although now Ethernet displaces all other networks, Taken-Ring can not be considered hopelessly outdated. More than 10 million computers around the world are combined with this network.

IBM has done everything for the widest possible dissemination of its network: detailed documentation has been released up to the adapter circuits. As a result, many companies, for example, 3som, Novell, Western Digital, Proteon and others started the production of adapters. By the way, the NetBIOS concept was developed specifically for this network, as well as for another IBM PC NetBiOS network. If the NetBIOS PC Network network has been kept in the NetBiOS-built-in permanent memory adapter, the NetBIOS emulation program has already been used on the TOKEN-Ring network. This allowed to respond more flexibly to the features of the equipment and maintain compatibility with higher level programs.

What is it - network technology? Why is it needed? What is used for? Answers to these, as well as a number of other issues and will be given under this article.

Several important parameters

1. Data transfer rate. This characteristic depends on what amount of information (measured in most cases in bits) can be transmitted via the network during a certain period of time.
2. Format frames. Information that is transmitted via the network is combined into information packages. They are called frames.
3. Signal coding type. In this case, it is solved, how to encrypt information in electrical impulses.
4. Transmission environment. Such a designation is used for the material, as a rule, this is a cable for which the flow of information flows is carried out, which is subsequent and displayed on the screens of monitors.
5. Topology network. This is a schematic constructing of the design that information transmits. Used, as a rule, tire, star and ring.
6. Access method.

A set of all these parameters and defines network technology than it is, which devices use and characteristics has. As you can guess, there are a great set.

general information

But what is the network technology? After all, the definition of this concept was never given! So, network technology is a coordinated set of standard protocols and hardware and hardware, which are implemented in a volume sufficient to build a local computer network. This determines how access to the data transfer medium will be obtained. Alternatively, you can still meet the name "Basic Technologies". To consider them all within the framework of the article is not possible due to a large amount, therefore attention will be paid to the most popular: Ethernet, token-ring, ArcNet and FDDI. What do they imagine?

Ethernet

At the moment it is the most popular network technology worldwide. If the cable is suggesting, the probability that it is used is, close to one hundred percents. Ethernet can safely be credited to the best network information technologies, which is due to low cost, high speed and quality of communication. The most famous is the IEEE802.3 / Ethernet type. But on its basis two very interesting options were developed. The first (IEEE802.3U / Fast Ethernet) allows you to provide a transfer rate of 100 Mbps. This option has three modifications. They differ among themselves using the material for the cable, the length of the active segment and the specific framework of the transmission range. But oscillations occur in the style of "plus-minus 100 Mbps". Another option is an IEEE802.3Z / Gigabit Ethernet. He has a transmitting ability to be 1000 Mbps. This variation has four modifications.

Token-Ring.

Network information technologies of this type are used to create a shared data transfer medium, which is eventually formed as a combination of all nodes into one ring. This technology is built on the star-ring topology. The first goes as the main one, and the second is extra. To access the network, the marker method is applied. The maximum ring length can be 4 thousand meters, and the number of nodes is 260 pieces. The data transfer rate does not exceed 16 Mbps.

Arcnet

This option uses the Tire and Passive Star Topology. At the same time, it can be built on a unshielded twisted pair and fiber optic cable. ArcNet is a real older network technology. The length of the network can reach 6000 meters, and the maximum number of subscribers is 255. In this case, it should be noted the main disadvantage of this approach - its low data transfer rate, which is only 2.5 Mbps / second. But this network technology is still widely used. This is due to its high reliability, low cost adapters and flexibility. Networks and network technologies built according to other principles may have higher speed indicators, but precisely because ArcNet provides high data output, it allows us to not throw it off from the accounts. An important advantage of this option is that the access method is used through the transfer of authority.

FDDI

Network computer technologies of this species are standardized high-speed data transmission architect specifications using fiber optic lines. The FDDI significantly affected ArcNet and Token-Ring. Therefore, this network technology can be considered as an improved data transfer mechanism based on available developments. The ring of this network can reach a hundred kilometers long. Despite the considerable distance, the maximum number of subscribers that can connect to it is only 500 knots. It should be noted that the FDDI is considered highly reliable due to the presence of basic and backup data routes. Adds her popularity and the ability to quickly transmit data - approximately 100 Mbps.

Technical aspect

Having considered that they represent the foundations of network technologies, which are used, now let's pay attention to how everything is arranged. Initially, it should be noted that the previously discussed options are exclusively local means of connecting electronics and computing machines. But there are global networks. All of them in the world about two hundred. How do modern network technologies work? To do this, let's consider the current principle of construction. So, there is a computer that are combined into one network. Conditionally they are divided into subscriber (basic) and auxiliary. The first are engaged in all informational and computational work. It depends on them, what will be the network resources. Auxiliary are engaged in transformation of information and its transmission over communication channels. Due to the fact that they have to handle a significant amount of data, servers boast of high power. But the final recipient of any information is still the usual host computer, which are most often represented by personal computers. Network information technologies can use such server types:

1. Network. Engaged in the transfer of information.
2. Terminal. Provides the functioning of the multiplayer system.
3. Databases. Engaged in handling database requests in multiplayer systems.

Channel switching networks

They are created thanks to the physical connection of customers at the time when messages are transmitted. What does it look in practice? In such cases, a direct connection is created to send and obtain information from point A to the point B. It includes channels of one of the set (as a rule) message delivery options. And the created connection for successful transmission must be unchanged during the entire session. But in this case, quite strong flaws are manifested. So, it is necessary to expect connections relatively long. This is accompanied by a high cost of data transmission and a low channel use ratio. Therefore, the use of network technologies of this type is not common.

Switching network messages

In this case, all information is transmitted in small portions. Direct connection in such cases is not established. Data transmission is carried out according to the first free of available channels. And so until the message is transmitted to its addressee. The servers are constantly engaged in the reception of information, its collection, verification and establishment of the route. And the subsequent message is transmitted further. Of the advantages it is necessary to note the low transmission price. But in this case, there are still problems such as low speed and the impossibility of the implementation of the dialogue between the AUM in real time.

Package switching networks

This is the most perfect and popular way to date. The development of network technologies led to the fact that now the exchange of information is carried out through short packages of information of the fixed structure. What do they imagine? Packages are part of messages that satisfy a specific standard. Small their length allows you to prevent network blocking. This reduces the queue in the switching nodes. A quick connection is carried out, a low level of errors is maintained, and considerable heights are achieved in terms of increasing the reliability and efficiency of the network. It should be noted that there are various configurations of this approach to the construction. So, if the network provides commutation of messages, packages and channels, then it is called integrated, that is, it is possible to hold it decomposition. Part of the resources can be used monopoly. So, some channels can be used to transmit direct messages. They are created at the time of data transfer between different networks. When a session of sending information ends, they disintegrate into independent trunk channels. When using batch technology, it is important to configure and negotiate a large number of customers, communication lines, servers and a number of other devices. This helps the establishment of rules that are known as protocols. They are part of the network operating system used and are implemented on hardware and programmatic levels.

Network technology local networks

In local networks, as a rule, a shared data transmission medium (monocanal) is used and the main role is assigned to protocols of the physical and channel levels, since these levels most reflect the specifics of local networks.

Network technology is a consistent set of standard protocols and implementing their software and hardware, sufficient to build a computing network. Network technologies are called basic technologies or network architectures.

Network architecture determines the topology and access method to the data transfer medium, cable system or data transmission medium, network frame format. Signal coding type, transmission rate. In modern computing networks, technologies or network architectures such as are: Ethernet, token-ring, ArcNet, FDDI received widespread.

IEEE802.3 / Ethernet Network Technologies

Currently, this architecture is most popular in the world. Popularity is provided by simple, reliable and inexpensive technologies. The classic Ethernet network uses a standard coaxial cable of two types (thick and thin).

However, the Ethernet version of the ethernet is becoming increasingly used as a medium of twisted pairs, since the installation and maintenance of them is much easier. In Ethernet networks, the Type Type and the Passive Star type are used, and the CSMA / CD access method.

The IEEE802.3 standard depending on the type of data transfer medium has modifications:

 10Base5 (thick coaxial cable) - provides the transfer rate of 10 Mbps and the length of the segment to 500m;

 10Base2 (thin coaxial cable) - provides the transfer rate of 10 Mbps and the length of the segment up to 200m ;;

 10Base-T (unshielded twisted pair) - allows you to create a network of star topology. The distance from the hub to the end node to 100m. The total number of nodes should not exceed 1024;

 10Base-F (fiber optic cable) - allows you to create a network of star topology. The distance from the hub to the end node to 2000m.
Ethernet technology has created high-speed options: IEEE802.3U / Fast Ethernet and IEEE802.3Z / Gigabit Ethernet. The main topology that is used in the networks Fast Ethernet and Gigabit Ethernet, passive star.

Network technology FAST Ethernet provides transmission speed of 100 Mbps and has three modifications:

 100Base-T4 - unshielded twisted pair (quad twisted pair) is used. Distance from the hub to the end node to 100m;

 100Base-TX - two twisted pairs are used (unshielded and shielded). Distance from the hub to the end node to 100m;

 100Base-FX - fiber optic cable (two fibers in the cable). The distance from the hub to the end node to 2000m; .

Gigabit Ethernet - Provides 1000 Mbps transmission rate. There are the following modifications of the standard:

 1000Base-SX - a fiber optic cable with an 850 nm light wavelength is used.

 1000Base-LX - a fiber optic cable with a light signal wavelength of 1300 nm is used.

 1000Base-CX - a shielded twisted pair is used.

 1000Base-T - a quadst unshielded twisted pair is used.
Fast Ethernet and Gigabit Ethernet Networks are compatible with networks made according to Ethernet standard, so easy and easy to connect Ethernet segments, Fast Ethernet and Gigabit Ethernet to a single computing network.

The only disadvantage of this network is the lack of a guarantee of the time of access to the environment (and mechanisms providing priority service), which makes the network is lowered to solve the technological tasks of real-time. Certain problems sometimes creates a limit on the maximum data field, equal to ~ 1500 bytes.

For different speed, Ethernet uses different encoding schemes, but the access algorithm and frame format remains unchanged, which guarantees software compatibility.

Ethernet frame has a format shown in fig.

Ethernet network frame format (numbers at the top of the figure show the size of the field in bytes)

Field preamble contains 7 bytes of 0haa and serves to stabilize and synchronize the medium (alternating CD1 and CD0 signals when completing CD0), follows the field SFD.START FRAME DELIMITER \u003d 0XAB), which is intended to identify the start of the frame. Field EFD. END FRAME DELIMITER) Specifies the end of the frame. Checksum field ( CRC -cyclic Redundancy Check), as well as preamble, SFD and EFD, are formed and monitored at the hardware level. In some protocol modifications, the EFD field is not used. The user is available fields starting with recipient addresses and ending the field informationinclusive. After CRC, an inter-package pause (IPG - Interpacket GAP is an interpasal interval) 9.6 μEK or more. The maximum frame size is equal to 1518 bytes (the preamble, SFD and EFD fields are not included here. The interface brings through all the packages following the cable segment to which it is connected, after all, to determine whether the package received is correct and to whom it is addressed only by taking it entirely. The correctness of the CRC package, in length and multiplicity, an integer byte is made after checking the destination address.

When the computer is connected to the network directly using the switch, the restriction on the minimum frame length is theoretically removed. But work with shorter personnel in this case will be possible only when replacing the network interface to non-standard (and, both at the sender and the recipient)!

If in the frame field protocol / Type Recorded code less than 1500, this field characterizes the length of the frame. Otherwise, this is the protocol code, the package of which is encapsulated in the Ethernet frame.

Access to the Ethernet channel is based on the algorithm CSMA / CD (Carrier Sense Multiple Access with Collision Detection).In Ethernet, any station connected to the network may try to start the packet transmission (frame) if the cable segment to which it is connected is free. Whether the segment is free, the interface determines the absence of a "carrier" for 9.6 μs. Since the first bit of the package reaches the rest of the network stations, it may happen that one or more stations will be attempted, especially since the delays in repeaters and cables can achieve sufficiently large values. Such coincidences of attempts are called clashes. The collision (collision) is recognized by the presence in the signal channel, the level of which corresponds to the operation of two or more transceivers simultaneously. When a collision is detected, the station interrupts transmission. The resumption of an attempt can be made after shutter speed (multiple of 51.2 mxkeg, but not exceeding 52 ms), the values \u200b\u200bof which is a pseudo-random value and is calculated by each station independently (T \u003d Rand (0.2 min (N, 10)), where N - The contents of the attempt counter, and the number 10 - backofflimit).

Usually, after a collision, the time is divided into a number of discrete domains with a double bus distribution period in the segment (RTT). For the maximum possible RTT, this time is equal to 512 bits. After the first collision, each station is waiting for 0 or 2 temporary domain before making another attempt. After a second collision, each of the stations may wait 0, 1, 2 or 3 temporary domain, etc. After n-collision, the random number lies within 0 - (2 n - 1). After 10 collisions, the maximum value of random excerpt ceases to grow and remains at 1023.

Thus, the longer cable segment, the greater the average access time.

After exposure, the station increases an attempt counter per unit and starts another transmission. The limit number of attempts by default is 16 if the number of attempts is exhausted, the connection is interrupted and the corresponding message is issued. The transmitted long frame contributes to the "synchronization" of the start of the transfer of packets by several stations. After all, during the transmission, with a noticeable probability, the need to transmit in two or more stations. At the time when they detect the completion of the package, the IPG timers will be included. Fortunately, information about the completion of the transfer of the package comes to the segment stations not at the same time. But the delays with which it is connected is also the reason that the fact of the beginning of the transfer of a new package of one of the stations does not become known immediately. When engaged in the collision of several stations, they may notify the remaining stations about it, sending the signal "ZATOR" (JAM is at least 32 bits). The contents of these 32 bits are not regulated. Such a scheme makes less likely re-collision. A source of a large number of collisions (in addition to the information overload) can serve as a proven total length of the logical cable segment, too large number of repeaters, cable breaking, no terminator (50-OMO Cable Corrector) or a malfunction of one of the interfaces. But the collision themselves are not something negative - this is a mechanism regulating access to the network environment.

In Ethernet, in the presence of synchronization, the following algorithms are possible:

BUT.

1. If the channel is free, the terminal transmits a package with a probability of 1.
2. If the channel is busy, the terminal is waiting for its release, after which the transfer is made.

B.

1. If the channel is free, the terminal transmits the package.
2. If the channel is busy, the terminal determines the time of the next transmission attempt. The time of this delay can be set in some statistical distribution.

IN.

1. If the channel is free, the terminal with the probability of P transmits the packet, and with the probability of 1-p, it deflares the transmission on T seconds (for example, to the next time domain).
2. When repetition attempts under the free channel, the algorithm does not change.
3. If the channel is busy, the terminal is waiting for the channel until free, after which it acts again according to the algorithm of paragraph 1.

The algorithm, and at first glance, it seems attractive, but it lates the possibility of collisions with a probability of 100%. Algorithms B and more resistant to this problem.

The effectiveness of the CSMA algorithm depends on how quickly the transmitting party learns about the fact of the collision and interrupts the transmission, because the continuation is meaningless - the data is already damaged. This time depends on the length of the network segment and delays in the equipment of the segment. A double delay value determines the minimum packet length transmitted to such a network. If the package is shorter, it can be transferred so that the transmitting party does not know about its damage due to the collision. For modern Local Ethernet networks built on switches and full-duplex connections, this problem is irrelevant

In order to explain this statement, consider the case when one of the stations (1) transmits the package of the remote computer itself (2) in this network segment. The time of distribution of the signal to this machine will be equal to T. Suppose also that the machine (2) will try to start the transmission just at the arrival of the package from the station (1). In this case, the station (1) learns about the collision only after time 2t after the start of the transmission (the signal propagation time from (1) to (2) plus the collision signal distribution time from (2) to (1)). It should be borne in mind that the register of collision is an analog process and the transmitting station should "listen" signal in the cable during the transmission process, comparing the read result with the fact that it transmits. It is important that the signal encoding scheme allow the collision detection. For example, the sum of two signals with a level of 0 this will not allow. You might think that the transfer of a short package with distortion due to the collision is not such a big trouble, the problem can solve the delivery and retransmission.

One should only consider that repeatedly transmitting in the case of a collision registered by the interface is carried out by the interface itself, and repeatedly transmitted in the case of the delivery control by the response is performed by the application process, requiring the resources of the central processor of the workstation.

Double turnover time and collision recognition

Clear recognition of collisions by all network stations is a prerequisite for the correct operation of the Ethernet network. If any transmitting station does not recognize the collisulation and decides that the data frame of it is faithful, then this data frame will be lost. Due to the applix of signals at the collision, the frame information will distort, and it will be discarded by the host station (possibly due to the discrepancy of the checksum). Most likely, distorted information will be re-submitted by any top-level protocol, such as transport or applied connections. But repeatedly transmitting the message protocols of the upper levels will occur after a significantly longer time interval (sometimes even after a few seconds) compared to microsecond intervals that the Ethernet protocol operates. Therefore, if the conflicts are not securely recognized by the Ethernet network nodes, this will lead to a noticeable reduction in the useful bandwidth of this network.

For reliable recognition of collisions, the following ratio should be performed:

T min\u003e \u003d pdv,

where T min - the transfer time of the frame of the minimum length, a PDV - the time for which the signal of the collision has time to spread to the long-range network node. Since in the worst case, the signal must go twice between the networks most remote from each other (a notcribed signal passes in one direction, and the signal is already distributed on the back path), then this time is called double revolutions time (Path Delay Value, PDV).

When this condition is fulfilled, the transmitting station should have time to detect the collision that caused the frame transferred to its frame, even before it finishes the transfer of this frame.

Obviously, the execution of this condition depends on the one hand, on the length of the minimum frame and the network bandwidth, and on the other hand, on the length of the cable system of the network and the speed of the signal propagation in the cable (for different types of cable, this speed is somewhat different).

All parameters of the Ethernet protocol are selected in such a way that with normal operation of the nodes of the collision network, the collision has always been clearly recognized. When the parameters are selected, of course, the relation given above is also taken into account, connecting the minimum length of the frame and the maximum distance between the stations in the network segment.

In the Ethernet standard, it is assumed that the minimum length of the data field of the frame is 46 bytes (which, together with service fields, gives the minimum length of the frame 64 bytes, and with the preamble - 72 bytes or 576 bits). From here it can be determined by the distance between the stations.

So, in the 10-megabit Ethernet, the minimum length transmission time is 575 bit intervals, therefore, the double turn time should be less than 57.5 μs. The distance that the signal can pass during this time depends on the type of cable and for a thick coaxial cable is approximately 13 280 m. Considering that during this time the signal must pass through the link twice, the distance between the two nodes should not be more than 6,635 m . In the standard, the value of this distance is essential less, taking into account other, more stringent restrictions.

One of these restrictions is associated with the maximum permissible attenuation of the signal. To ensure the required signal power during its passage between the maximum length of the cable segment, the maximum length of the continuous segment of a thick coaxial cable, taking into account the attenuation made by it, is selected 500 m. It is obvious that the Cable of 500 M Calculizive Conditions will be performed with a large reserve For frames of any standard length, including 72 bytes (double turnover of 500 m cable is only 43.3 bitted intervals). Therefore, the minimum length of the frame could be installed even less. However, technology developers did not reduce the minimum length of the frame, having in mind the multi-segment networks that are built from several segments connected by repeaters.

Repears increase the power transmitted from the segment to the signals segment, as a result, the attenuation of the signals is reduced and the network can be used a much larger length consisting of several segments. In coaxial implementations, Ethernet developers have limited the maximum number of segments on the network five, which in turn limits the total network length of 2500 meters. Even in such a multi-partition network, the detection condition of collisions is still performed with a large reserve (comparable from the permissible attenuation, a distance of 2500 m with a maximum possible signal to spread the signal with a distance of 6635 m). However, in reality, the time supply is significantly less, because in multi-segment networks, the repeaters themselves make an additional delay in several tens of bite intervals. Naturally, a small supply was also made to compensate for the deviations of cable parameters and repeaters.

As a result of the accounting of all these and some other factors, the ratio between the minimum frame and the maximum possible distance between the network stations was carefully selected between the network stations, which ensures reliable recognition of collisions. This distance is also called the maximum network diameter.

With increasing frame rate, which takes place in new standards based on the same CSMA / CD access method, such as Fast Ethernet, the maximum distance between the network stations is reduced in proportion to increasing the transmission rate. In the Fast Ethernet standard it is about 210 m, and in the Gigabit Ethernet standard it would be limited to 25 meters, if the standard developers did not take some measures to increase the minimum package size.

Calculation PDV.

To simplify the calculations, IEEE reference data is commonly used, containing the values \u200b\u200bof the delay of signals in repeaters, transceivers and various physical environments. In tab. 3.5 The data required to calculate the PDV value for all physical Ethernet network standards are given. The bit interval is indicated as BT.

Table 3.5. Data for calculating PDV value

The 802.3 committee tried to simplify the calculation as much as possible, therefore the data shown in the table includes several steps of signal passage. For example, the delays introduced by the repeater consist of a delay in the input transceiver, the delay of the repetition block and the delay of the output transceiver. Nevertheless, in the table, all these delays are represented by one value called the base of the segment. In order not to have been needed twice to fold the delays introduced by the cable, the tables are given twice values \u200b\u200bof delays for each type of cable.

The table uses such concepts as the left segment, the right segment and intermediate segment. Let us explain these terms on the example of the network shown in Fig. 3.13. The left segment is the segment in which the signal path from the transmitter exit (output in Fig. 3.10) of the final node begins. For example, this is a segment 1 . Then the signal passes through intermediate segments 2-5 and reaches the receiver (input R x in Fig. 3.10) the most remote node of the most remote segment 6, which is called right. It is here in the worst case a collision of frames occurs and conflict arises, which is meant in the table.

Fig. 3.13. An example of an Ethernet network consisting of segments of various physical standards

With each segment, a constant delay is associated with a base, which depends only on the type of segment and on the position of the segment on the signal path (left, intermediate or right). The base of the right segment in which the collision arises, much more exceeds the base of the left and intermediate segments.

In addition, the delay of the signal propagation along the segment cable is associated with each segment, which depends on the length of the segment and is calculated by multiplying the signal propagation time one by one cable meter (in bit intervals) on the length of the cable in meters.

The calculation is to calculate the delays introduced by each segment of the cable (the signal delay in the signal per 1 m in the table is multiplied by the length of the segment), and then the summation of these delays with the bases of the left, intermediate and right segments. The total PDV value should not exceed 575.

Since the left and right segments have different values \u200b\u200bof basic delay, in the case of any type of segments in the remote edges of the network must perform calculations twice: once to adopt as the segment of the left segment of the same type, and the second - a segment of another type. The result can be considered the maximum value of PDV. In our example, the extreme network segments belong to one type - standard 10Base-T, so it is not required double calculation, but if they were segments of different types, in the first case it would be necessary to take as the left segment between the station and the hub 1 and in the second to consider the left segment between the station and the concentrator 5 .

The network shown in the figure in accordance with the rule of 4 hubs is not correct - in the network between the segment nodes 1 and 6. There are 5 hubs, although not all segments are Lobase-FB segments. In addition, the total length of the network is 2800 m, which violates the rule of 2500 m. Calculate the value of PDV for our example.

Left segment 1 / 15.3 (base) + 100 * 0.113 \u003d 26.6.

Intermediate segment 2/ 33,5 + 1000 * 0,1 = 133,5.

Intermediate segment 3/ 24 + 500 * 0,1 = 74,0.

Intermediate segment 4/ 24 + 500 * 0,1 = 74,0.

Intermediate segment 5/ 24 + 600 * 0,1 = 84,0.

Right segment 6 /165 + 100 * 0,113 = 176,3.

The sum of all components gives the value of PDV, equal to 568.4.

Since PDV value less than the maximum permitted value 575, this network takes place by the criterion of time double turnover signal despite the fact that its total length is more than 2500 m, and the number of repeaters - more than 4

Calculation of PW.

To recognize the configuration of the network correctly, it is also necessary to calculate the decrease in the intercadron interval by repeaters, that is, the PW value.

To calculate PW also possible to use values \u200b\u200bof the maximum amounts of reducing inter-frame spacing when passing repeaters different physical media recommended by IEEE and given in the Table. 3.6.

Table 3.6. Reducing the intercadron interval by repeaters

In accordance with these data, calculate the PVV value for our example.

Left segment 1 10Base-T: Reduction at 10.5 BT.

Intermediate segment 2 10Base-FL: 8.

Intermediate segment 3 10Base-FB: 2.

Intermediate segment 4 10Base-FB: 2.

Intermediate segment 5 10Base-FB: 2.

The sum of these values \u200b\u200bgives the PW value of 24.5, which is less than the limit value of 49 bite intervals.

As a result, the network provided in the example complies with Ethernet standards in all parameters associated with the lengths of the segments, and with the number of repeaters

Maximum Ethernet Network Performance

The number of Ethernet processed frames per second is often indicated by bridge / switches and router manufacturers as the main characteristic of the performance of these devices. In turn, it is interesting to know the net maximum bandwidth Ethernet segment capacity in frames per second in the ideal case, when there is no network conflicts, and there are no additional delays introduced by bridges and routers. Such an indicator helps to evaluate the performance requirements for communication devices, since each port of the device cannot receive more frames per unit of time, which allows this to make the appropriate protocol.

For communication equipment, the most severe mode is the processing of frames of minimum length. This is because the processing of each frame bridge, switch or router spends about the same time associated with viewing table promotion package, the formation of a new frame (for the router), and so on. N. A number of minimum length frames entering the device per unit of time, naturally more than frames of any other length. Another characteristic of the performance of communication equipment - bit per second - is used less often, as it does not indicate what sized frames have processed the device, and on the maximum size frames to achieve high performance, measured in bits per second much easier.

Using the parameters shown in Table. 3.1, we calculate the maximum performance of the Ethernet segment in such units as the number of passes transmitted (packets) of the minimum length per second.

NOTE When you specify the network bandwidth, the term frame and a package are usually used as synonyms. Accordingly, the Frames-Per-Second, FPS and Packets-Per-Second, PPS, and PPS.

To calculate the maximum number of minimum length frames passing on Ethernet segment, we observe that the minimum frame length together with the preamble is 72 bytes or 576 bits (Fig. 3.5.), So its transmission is expended 57.5 microseconds. By adding an intercadron interval of 9.6 μs, we obtain that the period of reserves of the minimum length is 67.1 μs. From here the maximum possible bandwidth of the Ethernet segment is 14,880 frame / s.

Fig. 3.5. To the calculation of the bandwidth of the Ethernet protocol

Naturally, the presence in the segment of several nodes reduces this value by expecting access to the environment, as well as due to collisions leading to the need to re-transfer frames.

The maximum length frames of the Ethernet technology have a field of length of 1500 bytes, which, together with service information, gives 1518 bytes, and the preamble is 1526 bytes or 12,208 bits. The maximum possible bandwidth of the Ethernet segment for the frames of the maximum length is 813 frames / s. Obviously, when working with large frames, the load on bridges, switches and routers are quite significantly reduced.

Now we calculate what maximum useful bandwidth in bits per second have Ethernet segments when using the frames of different sizes.

Under useful bandwidth of the protocol It is understood as the transfer rate of user data that is transferred to the frame data field. This bandwidth is always less than the nominal bit speed of the Ethernet protocol due to several factors:

· service information frame;

· intercadron intervals (IPG);

· waiting for access to the environment.

For frames of minimum length, useful bandwidth is equal to:

With n \u003d 14880 * 46 * 8 \u003d 5.48 Mbps.

It is much less than 10 Mbps, but it should be noted that the minimum length frames are used mainly to transfer receipts, so that the transmission of the files actually does not have a relationship.

For frames of maximum length, useful bandwidth is equal to:

With n \u003d 813 * 1500 * 8 \u003d 9.76 Mbps,

which is very close to the rated speed of the protocol.

Once again we emphasize that such a speed can only be achieved when two interacting nodes in the Ethernet network other nodes do not interfere, which is extremely rare,

When using a medium-sized framework with a data field in 512 bytes, the network bandwidth will be 9.29 Mbps, which is also close enough to the limit bandwidth of 10 Mbps.

ATTENTION The ratio of the current network bandwidth to its maximum bandwidth is called network Utilization. At the same time, when determining the current bandwidth, the transfer of any information is taken into account on the network, both user and service. The coefficient is an important indicator for the technologies of shared media, since with a random character of the access method, the high value of the utilization ratio often speaks of a low utility network bandwidth (i.e., the speed of the user-defined bottom) - too much time the nodes spend on the procedure for obtaining access and repeated frames of frames After collisions.

In the absence of collisions and access to access, the network utilization ratio depends on the size of the frame data field and has a maximum value of 0.976 when transmitting frames of maximum length. Obviously, in the real Ethernet network, the average value of the network utilization can differ significantly from this value. More complex cases of determining the network bandwidth, taking into account the expectation of access and testing of collisions, will be discussed below.

Ethernet Personnel Formats

The Ethernet technology standard described in the IEEE 802.3 document gives a description of the single MAC level frame format. Since the LLC level frame, described in IEEE 802.2, should be seen in the MAC level frame, described in the IEEE document in the Ethernet standards, only the only channel of the channel-level frame can be used, the header of which is a combination of Mac headers and llc sublevels.

Nevertheless, in practice, ethernet networks use frames of 4 different formats (types). This is due to a long history of the development of the Ethernet technology, which has a period of existence before the adoption of IEEE 802 standards, when the LLC sublayer did not stand out from the general protocol and, accordingly, the LLC header was not applied.

In 1980, the consortium of three companies Digital, Intel and Xerox in 1980 submitted to the Committee 802.3 its branded version of the Ethernet standard (in which it was naturally described a certain format of the frame) as a project of an international standard, but the 802.3 committee adopted a standard characterized in some details from DIX offers. Differences related to the frame format, which generated the existence of two different types of frames in Ethernet networks.

Another format of the frame appeared as a result of Novell's efforts to speed up the work of its protocol stack in Ethernet networks.

Finally, the fourth format of the frame was the result of the activities of the Committee 802.2 to bring the previous framework formats to some general standard.

Differences in personnel formats can lead to incompatibility in the operation of equipment and network software, designed to work with only one Ethernet frame standard. However, today almost all network adapters, network adapter drivers, bridges / switches and routers can work with all Ethernet technology formats used in practice, and the frame type recognition is automatically executed.

The following is a description of all four types of Ethernet frames (here under the frame is understood as the entire set of fields that belong to the channel level, that is, the Mac and LLC levels). The same type of frame may have different names, so below for each type of frame, several most commonly used names are given:

· frame 802.3 / LLC (frame 802.3 / 802.2 or Frame Novell 802.2);

· frame RAW 802.3 (or Frame Novell 802.3);

· Ethernet DIX frame (or Ethernet II frame);

· Ethernet SNAP frame.

The formats of all these four types of Ethernet frames are shown in Fig. 3.6.

conclusions

· Ethernet is the most common technology of local networks. In the broad sense of Ethernet, this is a whole family of technology, including various branded and standard options, of which the most famous version of Ethernet DIX, 10-megabit variants of the IEEE 802.3 standard, as well as new high-speed Fast Ethernet and Gigabit Ethernet technologies. Almost all types of Ethernet technologies use the same method of separating the data transmission medium - the CSMA / CD random access method, which determines the appearance of technology as a whole.

· In the narrow sense of Ethernet - this is a 10 megabit technology described in the IEEE 802.3 standard.

· An important occurrence in Ethernet networks is the conflict - a situation where two stations simultaneously try to transfer the data frame on the general environment. The presence of collisions is an integral property of Ethernet networks, which is a consequence of an accepted random access method. The possibility of clearly recognition of collisions is due to the right choice of network parameters, in particular, the respect for the ratio between the minimum length of the frame and the maximum possible diameter of the network.

· On the performance characteristics of the network, the network utilization factor has great importance, which reflects its loading. With the values \u200b\u200bof this coefficient of over 50%, the useful bandwidth of the network drops sharply: due to the increase in the intensity of collisions, as well as an increase in the time of access to the environment.

· The maximum possible bandwidth of the Ethernet segment in frames per second is achieved when transferring frames of minimum length and is 14,880 frame / s. At the same time, the useful bandwidth of the network is only 5.48 Mbps, which is only slightly exceeding half of the nominal bandwidth - 10 Mbps.

· The maximum possible useful bandwidth of the Ethernet network is 9.75 Mbps, which corresponds to the use of frames of maximum length in 1518 bytes, which are transmitted over the network at a speed of 513 frames / s.

· In the absence of collisions and access in access coefficient of use The network depends on the size of the frame data field and has a maximum value of 0.96.

· Ethernet technology supports 4 different types of frames that have a common nodes format. There are formal signs by which network adapters automatically recognize the frame type.

· Depending on the type of physical environment, the IEEE 802.3 standard defines various specifications: 10Base-5, 10Base-2, 10Base-T, Foirl, 10Base-FL, 10Base-FB. For each specification, the type of cable is defined, the maximum lengths of the continuous segments of the cable, as well as the rules for using repeaters to increase the diameter of the network: the 5-4-3 rule for coaxial network options, and the 4-hubs rule for twisted pair and fiber.

· For a "mixed" network consisting of physical segments of various types, it is useful to calculate the total length of the network and the permissible number of repeaters. The IEEE 802.3 Committee provides initial data for such calculations, which indicate delays made by repeaters of various specifications of the physical environment, network adapters and cable segments.

Network Technologies IEEE802.5 / Token-Ring

TKEN Ring networks, as well as Ethernet network, characterizes a shared data transfer environment, which in this case consists of cable segments connecting all network stations in the ring. The ring is considered as a general shared resource, and it requires an accidental algorithm for access to it, but a deterministic, based on the transfer of the right to use the rings in a certain order. This right is transmitted using a special format frame called marker or token (Token).

TKEN Ring networks operate with two bit rates - 4 and 16 Mbps. Mixing stations operating at different speeds in one ring is not allowed. TKEN Ring networks operating at a rate of 16 Mbit / s have some improvements in the access algorithm compared to 4 Mbps standard.

TKEN Ring technology is more complex technology than Ethernet. It has the properties of fault tolerance. The token Ring network defines network control procedures that use the feedback of the ring-shaped structure - the sent frame is always returned to the station - the sender. In some cases, the detected errors in the operation of the network are eliminated automatically, for example, a lost marker can be restored. In other cases, errors are fixed only, and their elimination is performed by manually by service personnel.

To control the network, one of the stations acts on the role of the so-called active monitor. The active monitor is selected during the initialization of the ring as a station with the maximum value of the MAC address, if the active monitor fails, the ring initialization procedure is repeated and a new active monitor is selected. In order for the network to detect an active monitor failure, the last in the working condition every 3 seconds generates a special frame of its presence. If this frame does not appear on the network for more than 7 seconds, the rest of the network stations begin the procedure of elections of the new active monitor.

Frame formats TOKEN Ring

Token Ring exist three different frame formats:

· Marker;

· Data frame;

· interrupt sequence

Physical level TECKEN RING

Token Ring standard of IBM originally provided for building connections in the network using a hub, called MAU (Multistation Access Unit) or MSAU (Multi-Station Access Unit), ie multiple access devices (Fig. 3.15). TKEN Ring network may include up to 260 nodes.

Fig. 3.15. Physical Configuration Tank Ring Network

The TKEN Ring concentrator can be active or passive. The passive concentrator simply connects the ports internal connections so that the stations connecting to these ports form a ring. Neither the amplification of signals nor their resynication passive MSAU does not fulfill. Such a device can be considered a simple cross-block in one exception - MSAU ensures that the port is bypass when the computer attached to this port is turned off. This feature is necessary to ensure the connection of the ring, regardless of the state of the connected computers. Usually, the port bypass is performed due to the relay schemes that feed on a direct current from the network adapter, and when the network adapter is turned off, the relay is normal connected the port input to its output.

The active hub performs the signal regeneration functions and therefore is sometimes called the repeater, as in the Ethernet standard.

The question arises - if the hub is a passive device, how is the high-quality transmission of signals over long distances, which occur when the network is turned on into a network of several computers? The answer is that the role of the signal amplifier in this case takes on each network adapter, and the role of the resinchronizing unit is performed by the network adapter of the rings. Each TKEN Ring network adapter has a repetition unit that can regenerate and resynchronize signals, but the last function performs only the repetition unit of the active monitor in the ring.

The resynchronization unit consists of a 30-bit buffer, which takes manchester signals with a bit distorted during the turn around the ring of the reservation intervals. With the maximum number of stations in the ring (260), the variation of the battery circulation delay by the ring can reach 3-bit intervals. The active monitor "inserts" his buffer into the ring and synchronizes bit signals, giving them to the output with the required frequency.

In the general case, the TKen Ring network has a combined stellar-ring configuration. The end nodes are connected to the MSAU along the star topology, and the MSAU themselves are combined through special Ring In (Ri) and Ring Out (RO) ports for the formation of a trunk physical ring.

All stations in the ring should operate at one velocity - either 4 Mbps or 16 Mbps. Cables connecting the station with a hub, called branch (Lobe Cable), and cables connecting hubs - trunk (Trunk Cable).

TECKEN RING technology allows you to use various cable types for connecting end stations and hubs: STP Type I, UTP Type 3, UTP Type 6, as well as a fiber optic cable.

When using the shielded twisted pair of STP TYPE 1 from the IBM cable system nomenclature in the ring, it is allowed to combine up to 260 stations at the length of the branch cables up to 100 meters, and when using a unshielded twisted pair, the maximum number of stations is reduced to 72 at the length of the branch cables up to 45 meters.

The distance between the passive MSAU can reach 100 m when using cable STP Type 1 and 45 m when using UTP Type 3. Cable between active MSAU maximum distance increases correspondingly to 730 m or 365 m depending on the type of cable.

The maximum length of the TKEN Ring rings is 4000 m. Restrictions on the maximum ring length and the number of stations in the ring in the TOKEN RING technology are not so rigid as in Ethernet technology. Here, these restrictions are largely associated with the turnover of the marker along the ring (but not only - there are other considerations that dictate the choice of restrictions). Thus, if the ring consists of stations 260, then the token holding time of 10 ms token returns to the active monitor in the worst case through 2.6 seconds, and this time is just timeout control token rotation. In principle, all the values \u200b\u200bof the timeouts in the network adapters of the token Ring network adapters can be configured, so you can build a TKEN Ring network with a large number of stations and with a larger ring.

conclusions

· TKEN Ring technology is developing mainly by IBM and also has the status of the IEEE 802.5 standard, which reflects the most important improvements made to IBM technology.

· In TKEN Ring networks, a marker access method is used, which guarantees each station to gain access to the separated ring during the turnover time of the marker. Because of this property, this method is sometimes called deterministic.

· The access method is based on priorities: from 0 (lower) to 7 (higher). The station itself determines the priority of the current frame and can capture the ring only in the case when there are no more priority frames in the ring.

· Token Ring networks operate at two speeds: 4 and 16 Mbits / sec and can be used as the physical environment, shielded twisted pair, unshielded twisted pair, and optical fiber cable. The maximum number of stations in the ring - 260, and the maximum rings length is 4 km.

· TKEN Ring technology has elements of fault tolerance. Due to the feedback rings, one of the stations is an active monitor - continuously controls the presence of a marker, as well as the turnover time of the marker and data frames. If the ring is incorrectly, the procedure of re-initialization is launched, and if it does not help, the Beaconing procedure is used to locate a faulty cable or defective station.

· The maximum size of the data field of the TKEN Ring frame depends on the speed of the rings. For a speed of 4 Mbit / s, it is about 5,000 bytes, and at a speed of 16 Mbps - about 16 KB. The minimum size of the data field of the frame is not defined, that is, it can be 0.

· In the TKEN Ring network, the stations in the ring are combined with hubs called MSAU. The MSAU passive concentrator performs the role of a crossbar that connects the output of the previous station in the subsequent input ring. The maximum distance from the station to MSAU is 100 m for STP and 45 m for UTP.

· The active monitor performs in the ring also the role of the repeater - it resynchronizes the signals passing along the ring.

· The ring can be constructed on the basis of the active MSAU concentrator, which in this case is called the repeater.

· The TKEN Ring network can be based on a multiple rings separated by bridges routing frames on the principle "from the source", for which the special field is added to the TKEN Ring frame with the rings passing route.

Network technologies IEEE802.4 / ArcNet

As a topology, the ARCNET network uses the "tire" and "passive star". Supports shielded and unshielded twisted pair and fiber optic cable. In the ArcNet network, the method of transferring authority is used to access the data transmission medium. The ArcNet network is one of the oldest networks and has been very popular. Among the main advantages of the ARCnet network, you can call high reliability, low cost of adapters and flexibility. The main disadvantages of the network is the low speed of information transfer (2.5 Mbps). Maximum number of subscribers - 255. The maximum network length is 6000 meters.

Network Technology FDDI (Fiber Distributed Data Interface)

FDDI-standardized specification for high-speed data network architecture on fiber-optic lines. Transmission rate - 100 Mbps. This technology is largely based on the TECKE-RING architecture and uses deterministic marker access to the data transfer medium. The maximum length of the network rings is 100 km. The maximum number of network subscribers is 500. The FDDI network is a very highly reliable network that is created based on two fiber-optic rings forming the main and backup data transmission paths between nodes.

Main characteristics of technology

FDDI technology is largely based on TOKEN Ring technology, developing and improving its main ideas. The developers of the FDDI technology set themselves as the most priority objectives:

· increase the bit rate of data transfer up to 100 Mbps;

· increase the fault tolerance of the network due to standard procedures for restoring it after the failures of various kind - damage to the cable, incorrect operation of the node, hub, the occurrence of a high level of interference on the line, etc.;

· maximum effectively use the potential network bandwidth for asynchronous and synchronous (delay sensitive) traffic.

The FDDI network is based on two fiberglass rings that form the main and backup data transmission path between network nodes. The presence of two rings is the main way to improve fault tolerance in the FDDI network, and nodes that want to take advantage of this increased reliability potential must be connected to both rings.

In the normal network mode, the data passes through all the nodes and all sections of the cable only primary (PRIMARY) rings, this mode is called the mode Thru - "Through" or "transit". The secondary ring (SECONDARY) is not used in this mode.

In the event of a type of failure, when a part of the primary ring cannot transmit data (for example, a cable break or a node failure), the primary ring is combined with the secondary (Fig. 3.16), re-forming a single ring. This network mode is called Wrap, That is, "coagulation" or "folding" rings. The coagulation operation is performed by means of hub and / or FDDI network adapters. To simplify this procedure, data on the primary ring is always transmitted in one direction (on the diagrams this direction is depicted counterclockwise), and on the secondary - in the opposite (depicted clockwise). Therefore, when the common ring of two rings is formed, the stations transmitters still remain connected to the receivers of adjacent stations, which allows you to correctly transmit and receive information to neighboring stations.

Fig. 3.16. Reconfiguration of FDDI rings when refusing

In FDDI standards, much attention is paid to various procedures, which allow you to determine the presence of a network failure, and then produce the necessary reconfiguration. The FDDI network can fully restore its performance in the event of single failures of its elements. With multiple failures, the network disintegrates into several non-related networks. The FDDI technology complements the mechanisms for detecting the token Ring technology technology mechanisms for the reconfiguration of data transfer paths in a network based on the presence of reserve connections provided by the second ring.

Rings in FDDI networks are treated as a common dividing data environment, so it defines a special access method. This method is very close to the TKEN Ring network access method and is also called the TKEN Ring ring.

The differences between the access method are that the retaining time of the marker on the FDDI network is not a permanent value as in the TECK Ring network. This time depends on the loading of the ring - with a small load it increases, and with large overloads it can decrease to zero. These changes in the access method concern only asynchronous traffic that is not critical to small frame transfer delays. For synchronous traffic, the retaining time of the marker is still a fixed value. The mechanism of personnel priorities, similar to the TECK Ring adopted in technology, is absent in the FDDI technology. Technology developers decided that the division of traffic on 8 levels of priorities is redundant and sufficiently split traffic into two classes - asynchronous and synchronous, the last of which is always serviced, even when the ring overloads.

Otherwise, forwarding frames between the MAC Ring Stations is fully consistent with TKen Ring technology. FDDI stations use the marker early release algorithm, like TKen Ring network at a speed of 16 Mbps.

MAC level addresses have a standard format for IEEE 802 technologies. FDDI frame format is close to the format of the Frame Taken Ring, the main differences are in the absence of priority fields. Signs of address recognition, frame copying and error allow you to save on token Ring network processing procedures in the sender station, intermediate stations and a recipient station.

In fig. 3.17 This correspondence for the structure of the FDDI technology of the seven-level OSI model. FDDI Specifies the physical layer protocol and an access control system (Mac) of the channel layer. As in many other local network technologies, the FDDI technology uses the LLC data control of the LLC data control system defined in the IEEE 802.2 standard. Thus, despite the fact that the FDDI technology was developed and standardized by the ANSI Institute, and not the IEEE committee, it fully fits into the structure of 802 standards.

Fig. 3.17. FDDI technology protocol structure

A distinctive feature of the FDDI technology is the station management level - Station Management (SMT). It is the SMT level that performs all functions for managing and monitoring all other levels of the FDDI protocols stack. Each FDDI network node takes part in the rings. Therefore, all nodes exchange SMT special samples to control the network.

FDDi network failover is provided by protocols and other levels: Network failures are eliminated by physical levels for physical reasons, for example, due to the cable break, and using the MAC level - logical network failures, for example, the loss of the desired internal path of the marker transmission and data frames between the concentrator ports .

conclusions

· The FDDI technology was first used a fiber optic cable in local networks, as well as operation at a speed of 100 Mbps.

· There is a significant continuity between TECKE Ring and FDDI technologies: both ring topology and a marker access method are characterized.

· FDDI technology is the most fault tolerant technology of local networks. With one-time failures of the cable system or station, the network, due to the "folding" of the double ring to the single, remains quite efficient.

· A marker FDDI access method works differently for synchronous and asynchronous frames (frame type defines station). To transmit a synchronous frame, the station can always capture the coming marker at a fixed time. To transmit asynchronous frame, the station can capture the marker only when the marker performed turnover over the ring quickly, which indicates the absence of ring overloads. Such an access method, firstly, prefers synchronous frames, and secondly, regulates the loading of the ring, slowing down the transmission of irregular asynchronous frames.

· As a physical environment, the FDDI technology uses fiber optic cables and UTP category 5 (this version of the physical layer is called TP-PMD).

· The maximum number of double-connected stations in the ring - 500, the maximum dual ring diameter is 100 km. The maximum distances between adjacent nodes for the multimode cable are 2 km, for twisted UPT pairs of category 5-100 m, and for single-mode fiber depends on its quality.

Computer networks are divided into three main classes:

1. Local computer networks (LAN - LocalaReaNetwork) are networks that combine computers that are geographically in one place. The local network combines computers located physically close to each other (in one room or one building).

2. Regional computer networks (MAN - MetropolitanareAnetwork) are networks that combine several local computer networks located within one territory (city, regions or region, for example, the Far East).

3. Global Computing Networks (WAN - WideareAnetwork) are networks that combine many local, regional networks and

computers of individual users located at any distance from each other (Internet, Fido).

The following standards for building local computing networks are currently used:

ArcNet; (IEEE 802.4)

Token Ring; (802.5)

Ethernet. (802.3)

Consider each of them Read more

TechnologyIeee 802.4 ArcNet (or ArcNet, from English. Attached Resource Computer Network) - LAN technology, the purpose of which is similar to the purpose of Ethernet or TKEN Ring. ArcNet was the first technology to create microcomputer networks and became very popular in the 1980s in the automation of the institution. Designed for the organization LAN in the network topology "Star".

The basis of communication equipment is:

switch (Switch)

passive / Active hub

The advantage has switting equipment, as it allows you to form network domains. Active hubs are applied with a large removal of the workstation (they restore the signal shape and enhance it). Passive - with small. The network uses the assignable principle of workstation access, that is, the right to transfer the station that received the so-called software marker from the server. That is, deterministic network traffic is implemented.

Advantages of the approach:

Remarks: Messages transmitted by workstations form a queue on the server. If the queue maintenance time is significantly (more than 2 times) exceeds the maximum package delivery time between the two most remote stations, it is believed that the network bandwidth reached the maximum limit. In this case, further network extension is impossible and the installation of the second server is required.

Limit technical characteristics:

The minimum distance between workstations connected to one cable is 0.9 m.

The maximum length of the network along the longest route is 6 km.

Restrictions are associated with hardware retention of information transfer with a large number of commuting elements.

The maximum distance between the passive concentrator and the workstation is 30 m.

The maximum distance between the active and passive hub is 30 m.

Between active hub and active hub - 600 m.

Advantages:

Low cost of network equipment and the ability to create extended networks.

Disadvantages:

Low data transfer rate. After the distribution of Ethernet as technology to create a LAN, ArcNet has been used in embedded systems.

The support of ArcNet technology (in particular the distribution of specifications) is engaged in the non-profit organization ArcNet Trade Association (ATA).

Technology - ArcNet architecture is represented by two main topologies: tire and star. As a transmission medium, the RG-62 coaxial cable with a 93 ohm wave resistance is used, a plug-based plugs with an appropriate seal diameter (differ from the forks 10Base-2 (thin Ethernet)).

Network equipment consists of network adapters and hubs. Network adapters can be for tire topology, for star and universal. Hubs can be active and passive. Passive hubs are applied to create stellar networks. Active hubs can be for tire, star and mixed topology. The ports for tire topology are physically not compatible with ports for star topology, although they have the same physical connection (BNC socket).

In the case of tire topology, workstations and servers are connected to each other using T connectors (the same as in 10Base-2 (thin "Ethernet)) connected to network adapters and hubs and connected coaxial cable. The extreme points of the segment are terminated with tips with a resistance of 93 ohms. The number of devices on one bus is limited. The minimum distance between the connectors is 0.9 meters and should be more than this magnitude. To facilitate the cutting, labels can be applied to the cable. Separate tires can be combined with tire hubs.

When using star topology, active and passive hubs are applied. Passive hub is a resistive summitor that allows you to connect four cables. All cables in this

the case is connected according to the "point-to-point" principle, without the formation of tires. Between the two active devices, more than two passive hubs should not be connected. The minimum length of any network cable is 0.9 meters and should be a multiple of this magnitude. There is a limitation of the cable length between active and passive ports, between two passive, between two active.

With mixed topology, active hubs are used that support both types of connectivity.

On network adapters of workstation and servers using jumpers or DIP switches, a unique network address is set to the use of the BIOS extension chip, which allows you to remotely boot the workstation (may be disusser), connection type (tire or stellar topology), connecting the built-in terminator ( The last two points is optional). Restriction on the number of workstations - 255 (on the discharge of a network address register). In case two devices have the same network address, both lose their performance, but this conflict does not affect the operation of the network.

With a bus topology, the cable closer or terminator leads to the non-working capacity of the network for all devices connected to the segment, which includes this cable (that is, from the terminator to the terminator). With a starry topology, the breakdown of any cable leads to the failure of the segment, which is turned off by this cable from the server file.

ArcNet logical architecture - ring with marker access. Since such an architecture, in principle, does not allow conflicts, with a relatively large number of hosts (in practice 25-30 workstations was tested) The performance of the ArcNet network was higher than 10Base-2, with four times less in the medium (2.5 versus 10 Mbps ).

TECHNOLOGY 802.5 TOKEN RING - Local computing network technology (LAN) Rings with "marker access" - a local network protocol, which is located on a channel level (DLL) of the OSI model. It uses a special three-seater frame called a marker that moves around the ring. The ownership of the marker provides the right to the owner to transfer information on the carrier. The frames of the ring network with markers are moved to the cycle. In the local computing network (LAN) TOKEN Ring are logically organized into a ring topology with data transmitted sequentially from one ring station to another with a control marker circulating around ring access control. This marker transmission mechanism is shared by ArcNet, a marker tire, and FDDI, and has theoretical advantages over stochastic CSMA / CD Ethernet.

Marker's transfer token Ring and IEEE 802.5 are the main examples of networks with marker. The network with the transmission of the marker is moved along the network a small data block called marker. Owning this marker guarantees the right to transfer. If the node accepting the marker does not have information for sending, it simply rewinds the marker to the next end station. Each station can hold the marker for a certain maximum time (by default - 10 ms).

This technology offers an option to solve the problem of collisions, which occurs when the local network is operating. In Ethernet technology, such conflicts occur while simultaneously transmitting information to several workstations located within one segment, that is, using a common physical data channel.

If the station that owns the marker, there is information for transmission, it captures the marker, it changes in his one bit (as a result of which the marker turns into a "beginning of the data block" sequence), complements the information that he wants to transfer and sends this information to the next station ring network. When the information block circulates over the ring, there is no marker in the network (if only the ring does not provide an "early release of the marker" - Early Token Release), so other stations wishing to transfer information to expect. Consequently, in networks token Ring can not be conflicts. If the earlier release of the marker is provided, the new marker can be released after completing the transmission of the data block.

The information block circulates over the ring until it reaches the intended destination station, which copies the information for further processing. The information block continues to circulate by ring; It is finally removed after reaching the station, which fishes this unit. The sending station can check the returned block to make sure it has been viewed and then copied by the destination station.

The scope of application in contrast to CSMA / CD networks (for example, Ethernet) networks with marker transfer are deterministic networks. This means that you can calculate the maximum time, which will pass before any end station can be transmitted. This feature, as well as some reliability characteristics, make the network token Ring ideal for applications where the delay must be predictable and the stability of the network is important. Examples of such applications is the environment of automated stations on the factories.

It is used as a cheaper technology, gained distribution everywhere where there are responsible applications for which not so much speed as reliable delivery of information is important. Currently, Ethernet for reliability is not inferior to token Ring and significantly higher in performance.

Modifications token RingSupply 2 modifications in transmission speeds: 4 Mbps and 16 Mbps. In Token Ring 16 Mbps used

an early liberation technology of the Marker. The essence of this technology is that the station, "capturing" marker, is generated by a free marker at the end of data transfer and launches it to the network. Attempts to implement 100 Mbps did not be crowned with commercial success. Currently, TKen Ring technology is not supported.

802.3 Ethernet technology from English. Ether "Ether") - Package technology of data transmission mainly local computer networks.

Ethernet standards define wired connections and electrical signals at the physical level, frame format and medium access control protocols - on channel level of the OSI model. Ethernet is mainly described by group 802.3. Ethernet has become the most common LAN technology in the mid-90s of the last century, proving such outdated technologies such as ArcNet, FDDI and TKEN Ring.

The following should be taken into account on the creation of a local network:

* Creating a local network and setup equipment for accessing the Internet;

* The choice of equipment should be based on the specifications capable of meeting the requirements for the data transfer rate;

* The equipment must be safe, protected from electric shock;

* Each workstation has a network cable to connect to the network;

* Possible Wi-Fi throughout the Account;

* The location of jobs should satisfy the requirements of standards for placing equipment in educational institutions;

* The cost of creating a local network should be economically justified;

* Local network reliability.