Let's note the main features of the development of Ethernet networks and the transition to networks Fast Ethernet(IEEE 802.3u standard):

• - tenfold increase in throughput;
• - preservation of the random access method CSMA / CD;
• - preservation of the frame format;
• - support for traditional data transmission media.

These features, as well as support for two speeds and auto-sensing 10/100 Mbps, built into network cards and Fast Ethernet switches, allow for a smooth transition from Ethernet networks to faster Fast Ethernet networks, providing an advantageous succession compared to other technologies. Another additional factor for successful market penetration is the low cost of Fast Ethernet equipment.

Fast Ethernet architecture

Fast Ethernet layer structure (including MII interface and Fast Ethernet transceiver) is shown in Fig. 13. At the stage of developing the 100Base-T standard, the IEEE 802.3u committee determined that there is no universal signal coding scheme that would be ideal for all three physical interfaces (TX, FX, T4). Compared to the Ethernet standard, the encoding function (Manchester code) is performed by the physical signaling layer PLS (Fig. 5), which is located above the medium-independent interface AUI. In the Fast Ethernet standard, the encoding functions are performed by the PCS encoding sublayer located below the medium-independent MII interface. As a result, each transceiver must use its own set of coding schemes, the best way suitable for the corresponding physical interface, such as a 4V / 5V kit and NRZI for a 100Base-FX interface.

MII interface and Fast Ethernet transceivers. MII (medium independent interface) in Fast Ethernet is analogous to AUI in Ethernet. The MII interface provides communication between the negotiation and physical coding sublayers. Its main purpose is to make it easier to use. different types Wednesday. MII interface assumes further connection of Fast Ethernet transceiver. A 40-pin connector is used for communication. The maximum distance over the MII interface cable should not exceed 0.5 m.

If the device has standard physical interfaces (for example, RJ-45), then the structure of the physical layer sublayers can be hidden inside the microcircuit with a large logic integration. In addition, deviations in the protocols of intermediate sublevels in a single device are permissible, with the main goal of increasing performance.

Physical interfaces Fast Ethernet

The Fast Ethernet IEEE 802.3u standard establishes three types of physical interface (Fig. 14, Table 6 Main characteristics of physical interfaces of the IEEE 802.3u Fast Ethernet standard): 100Base-FX, 100Base-TX and 100Base-T4.

100Base-FX. The standard for this fiber optic interface is completely identical to the FDDI PMD standard. The main optical connector of the 100Base-FX standard is Duplex SC. The interface allows a full duplex communication channel.

• * - the distance is reached only in duplex communication mode.
• 100Base-TX. The standard for this physical interface assumes the use of an unshielded twisted pair of category 5 or higher. It is completely identical to the FDDI UTP PMD standard. The physical RJ-45 port, as in the 10Base-T standard, can be of two types: MDI (network cards, workstations) and MDI-X (Fast Ethernet repeaters, switches). A single MDI port may be present on a Fast Ethernet repeater.

Pairs 1 and 3 are used for transmission over copper cable. Pairs 2 and 4 are free. The RJ-45 port on the network card and on the switch can support, in addition to the 100Base-TX mode, the 10Base-T mode, or the auto-sensing function. Most modern network cards and switches support this function over RJ-45 ports and, in addition, can operate in full duplex mode.

100Base-T4. This type of interface allows providing a half-duplex communication channel over a twisted pair UTP cat. 3 and higher. It is the ability of an enterprise to migrate from Ethernet to Fast Ethernet without radical replacement of the existing cabling system based on UTP cat.3 that should be considered the main advantage of this standard.

Unlike the 100Base-TX standard, where only two twisted pairs of cable are used for transmission, the 100Base-T4 standard uses all four pairs. Moreover, when communicating workstation and the repeater through a direct cable, data from the workstation to the repeater goes through twisted pairs 1, 3 and 4, and in the opposite direction - through pairs 2, 3 and 4, Pairs 1 and 2 are used for collision detection like the Ethernet standard. The other two pairs 3 and 4 alternately, depending on the commands, can pass the signal either in one direction or in the other direction. Transmitting a signal in parallel over three twisted pairs is equivalent to inverse multiplexing discussed in Chapter 5. The bit rate per channel is 33.33 Mbps.

Character coding 8B / 6T... If Manchester coding were used, the bit rate per twisted pair would be 33.33 Mbit / s, which would exceed the established 30 MHz limit for such cables. An effective reduction in the modulation frequency is achieved by using a ternary code instead of a direct (two-level) binary code. This code is known as 8B / 6T; this means that before transmission takes place, each set of 8 binary bits (character) is first converted according to certain rules into 6 triple (three-level) characters.

The 100Base-T4 interface has one significant drawback - the fundamental impossibility of supporting the duplex transmission mode. And if during the construction of small Fast Ethernet networks using 100Base-TX repeaters it has no advantages over 100Base-T4 (there is a collision domain, the bandwidth of which is not more than 100 Mbit / s), then during the construction of networks using switches, the lack of the 100Base-T4 interface becomes obvious and very serious. Therefore, this interface is not as widespread as 100Base-TX and 100Base-FX.

Fast Ethernet device types

The main categories of devices used in Fast Ethernet are the same as in Ethernet: transceivers; converters; network cards (for installation on workstations / file servers); repeaters; switches.

Transceiver- a two-port device covering the PCS, PMA, PMD and AUTONEG sublevels, and having, on the one hand, an MII interface, on the other, one of the environment-dependent physical interfaces (100Base-FX, 100Base-TX or 100Base-T4). Transceivers are used relatively rarely, as well as network cards, repeaters, switches with an MII interface are rarely used.

LAN card. The most widespread today are network cards with a 100Base-TX interface to the PCI bus. Optional but highly desirable features of the RJ-45 port are 100/10 Mbps autoconfiguration and full duplex support. Most of the current cards being released support these features. There are also network cards with 100Base-FX optical interface (manufacturers IMC, Adaptec, Transition Networks, etc.) - the main standard optical connector is SC (ST allowed) for multimode fiber.

Converter(media converter) - a two-port device, both ports of which represent media-dependent interfaces. Converters, unlike repeaters, can operate in full duplex mode, except when there is a 100Base-T4 port. 100Base-TX / 100Base-FX converters are widespread. Due to the general growth trends of long-distance broadband networks using single-mode FOC, consumption optical transceivers on a single-mode optical fiber has increased sharply in the last decade. Converter chassis that combine multiple separate 100Base-TX / 100Base-FX modules allow multiple converging fiber segments to be connected to a switch equipped with RJ-45 (100Base-TX) duplex ports.

Repeater. According to the parameter of maximum time delays during frame retransmission, Fast Ethernet repeaters are divided into two classes:

• - Class I. Delay on double run RTD should not exceed 130W. Due to less stringent requirements, repeaters of this class can have T4 and TX / FX ports, and can also be stacked.
• - Class II. Repeaters of this class have more stringent double run delay requirements: RTD

Switch- an important device corporate networks... Most modern Fast Ethernet switches support 100/10 Mbit / s auto-configuration over RJ-45 ports and can provide full duplex communication on all ports (except for 100Base-T4). Switches can have special additional slots for installing an up-link module. Optical ports such as Fast Ethernet 100Base-FX, FDDI, ATM (155 Mbit / s), Gigabit Ethernet, etc. can act as interfaces for such modules.

Large switch manufacturers Fast Ethernet companies are: 3Com, Bay Networks, Cabletron, DEC, Intel, NBase, Cisco, etc.

The most widespread among standard networks is the Ethernet network. It first appeared in 1972 (developed by the well-known company Xerox). The network turned out to be quite successful, and as a result of this, in 1980 it was supported by such major companies as DEC and Intel (the merger of these companies was named DIX after the first letters of their names). Through their efforts in 1985, the Ethernet network became an international standard, it was adopted by the largest international standards organizations: the 802 IEEE committee (Institute of Electrical and Electronic Engineers) and ECMA (European Computer Manufacturers Association).

The standard was named IEEE 802.3 (in English it reads as eight oh two dot three). It defines multiple access to a mono-channel of the bus type with collision detection and transmission control, that is, with the already mentioned CSMA / CD access method. Some other networks also met this standard, since the level of detail is low. As a result, networks of the IEEE 802.3 standard were often incompatible with each other in terms of both design and electrical characteristics. Recently, however, the IEEE 802.3 standard is considered the standard for the Ethernet network.

Key features of the original IEEE 802.3 standard:

• topology - bus;
• transmission medium - coaxial cable;
• transmission speed - 10 Mbit / s;
• maximum network length - 5 km;
• the maximum number of subscribers is up to 1024;
• network segment length - up to 500 m;
• number of subscribers on one segment - up to 100;
• access method - CSMA / CD;
• transmission is narrowband, that is, without modulation (mono channel).

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

Ethernet is now the most popular in the world (over 90% of the market) and it is expected to remain so in the coming years. This was largely facilitated by the fact that from the very beginning the characteristics, parameters, protocols of the network were open, as a result of which a huge number of manufacturers around the world began to produce Ethernet equipment that is fully compatible with each other.

In the classic Ethernet network, a 50-ohm coaxial cable of two types (thick and thin) was used. However, in recent years (since the beginning of the 90s), the most widespread version of Ethernet is using twisted pairs as a transmission medium. A standard has also been defined for the use of fiber optic cable in a network. Additions have been made to the original IEEE 802.3 standard to accommodate these changes. In 1995, an additional standard appeared for a faster version of Ethernet operating at a speed of 100 Mbit / s (the so-called Fast Ethernet, IEEE 802.3u standard), using twisted pair or fiber-optic cable as the transmission medium. In 1997, a version with a speed of 1000 Mbit / s appeared (Gigabit Ethernet, IEEE 802.3z standard).

In addition to the standard bus topology, passive star and passive tree topologies are increasingly being used. This assumes the use of repeaters and repeater hubs connecting different parts (segments) of the network. As a result, a tree-like structure can be formed on segments of different types (Fig. 7.1).

Fig. 7.1. Classic Ethernet topology

The segment (part of the network) can be a classic bus or a single subscriber. The bus segments use coaxial cable, and the passive star beams (for connecting single computers to the hub) use twisted pair and fiber optic cables. The main requirement for the resulting topology is that there are no closed paths (loops) in it. In fact, it turns out that all subscribers are connected to a physical bus, since the signal from each of them propagates in all directions at once and does not return back (as in a ring).

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

Fast Ethernet does not have a physical bus topology, only a passive star or passive tree is used. In addition, Fast Ethernet has much more stringent requirements for the maximum network length. Indeed, when the transmission speed is increased by 10 times and the format of the packet is preserved, its minimum length becomes ten times shorter. Thus, the permissible value of the double signal transit time through the network decreases by 10 times (5.12 μs versus 51.2 μs in Ethernet).

The standard Manchester code is used to transmit information on the Ethernet network.

Access to the Ethernet network is carried out using a random CSMA / CD method, which ensures the equality of subscribers. The network uses packets of variable length with the structure shown in Fig. 7.2. (numbers show the number of bytes)

Fig. 7.2. Ethernet packet structure

The Ethernet frame length (that is, the packet without the preamble) must be at least 512 bit intervals or 51.2 µs (this is the limit for double network transit time). Provides individual, multicast and broadcast addressing.

The Ethernet packet includes the following fields:

• The preamble consists of 8 bytes, the first seven are the code 10101010, and the last byte is the code 10101011. In the IEEE 802.3 standard, the eighth byte is called the Start of Frame Delimiter (SFD) and forms a separate field of the packet.
• The recipient (receiver) and sender (transmitter) addresses are 6 bytes each and are built according to the standard described in the Packet Addressing section of Lecture 4. These address fields are processed by the subscribers' equipment.
• The control field (L / T - Length / Type) contains information about the length of the data field. It can also determine the type of protocol used. It is generally accepted that if the value of this field is not more than 1500, then it indicates the length of the data field. If its value is more than 1500, then it determines the frame type. The control field is processed programmatically.
• The data field must contain between 46 and 1500 bytes of data. If the packet is to contain less than 46 bytes of data, then the data field is padded with padding bytes. According to the IEEE 802.3 standard, a special padding field (pad data) is allocated in the packet structure, which can have zero length when there is enough data (more than 46 bytes).
• The Frame Check Sequence (FCS) field contains a 32-bit cyclic packet checksum (CRC) and is used to check the correctness of the packet transmission.

Thus, the minimum frame length (packet without preamble) is 64 bytes (512 bits). It is this value that determines the maximum allowable double delay of signal propagation over the network in 512 bit intervals (51.2 μs for Ethernet or 5.12 μs for Fast Ethernet). The standard assumes that the preamble may shrink as the packet passes through various network devices, so it is ignored. The maximum frame length is 1518 bytes (12144 bits, i.e. 1214.4 μs for Ethernet, 121.44 μs for Fast Ethernet). This is important for sizing buffer memory network equipment and to assess the overall network load.

The choice of the preamble format is not accidental. The point is that the sequence of alternating ones and zeros (101010 ... 10) in the Manchester code is characterized by the fact that it has transitions only in the middle of the bit intervals (see Section 2.6.3), that is, only information transitions. Of course, it is easy for the receiver to tune (synchronize) with such a sequence, even if for some reason it is shortened by a few bits. The last two unit bits of the preamble (11) differ significantly from the sequence 101010 ... 10 (transitions also appear at the border of the bit intervals). Therefore, the already tuned receiver can easily select them and thereby detect the beginning of useful information (the beginning of the frame).

For an Ethernet network operating at a speed of 10 Mbit / s, the standard defines four main types of network segments, focused on different media of information transmission:

• 10BASE5 (thick coaxial cable);
• 10BASE2 (thin coaxial cable);
• 10BASE-T (twisted pair);
• 10BASE-FL (fiber optic cable).

The segment name includes three elements: the number 10 means the transmission rate of 10 Mbit / s, the word BASE means transmission in the main frequency band (that is, without modulation of the high-frequency signal), and last element- permissible segment length: 5 - 500 meters, 2 - 200 meters (more precisely, 185 meters) or the type of communication line: T - twisted pair (from English twisted-pair), F - fiber optic cable (from English fiber optic).

Likewise, for an Ethernet network operating at a speed of 100 Mbps (Fast Ethernet), the standard defines three types of segments, differing in the types of transmission media:

• 100BASE-T4 (twisted pair);
• 100BASE-TX (twisted pair);
• 100BASE-FX (fiber optic cable).

Here, the number 100 stands for a transmission rate of 100 Mbps, the letter T for a twisted pair, and the letter F for a fiber optic cable. The 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 features of Ethernet equipment, as well as the CSMA / CD exchange control algorithm and the cyclic checksum (CRC) calculation algorithm will be discussed in more detail later in special sections of the course. It should be noted here only that the Ethernet network does not differ in either record characteristics or optimal algorithms; it is inferior in a number of parameters to other standard networks. But thanks to its powerful support, the highest level of standardization, huge volumes of production of technical means, Ethernet favorably stands out among other standard networks, and therefore it is customary to compare any other network technology with Ethernet.

The evolution of Ethernet technology is moving away from the original standard. The use of new transmission media and switches can significantly increase the size of the network. Abandoning the Manchester code (on Fast Ethernet and Gigabit Ethernet) results in higher data rates and reduced cable requirements. Rejection of the CSMA / CD control method (with full-duplex exchange mode) makes it possible to dramatically increase the efficiency of work and remove restrictions on the length of the network. However, all of the newer varieties of networking are also referred to as Ethernet.

Token-Ring network

The Token-Ring (token ring) network was proposed by IBM in 1985 (the first option appeared in 1980). It was designed to network all types of computers made by IBM. The very fact that it is supported by IBM, the largest manufacturer of computer equipment, suggests that it needs to be given special attention. But no less important is the fact that Token-Ring is currently the international standard IEEE 802.5 (although there are minor differences between Token-Ring and IEEE 802.5). This puts this network one level in status with Ethernet.

Developed by Token-Ring as a reliable alternative to Ethernet. Although Ethernet is now superseding all other networks, Token-Ring is not hopelessly obsolete. More than 10 million computers worldwide are connected by this network.

IBM has done everything to distribute its network as widely as possible: detailed documentation was released up to schematic diagrams adapters. As a result, many companies, for example, 3COM, Novell, Western Digital, Proteon and others, started to manufacture adapters. By the way, the NetBIOS concept was developed specifically for this network, as well as for another IBM PC Network. Whereas in the previously created PC Network, NetBIOS programs were stored in the read-only memory built into the adapter, in the Token-Ring network, a NetBIOS emulation program was already used. This made it possible to more flexibly respond to the peculiarities of the hardware and maintain compatibility with higher-level programs.

The Token-Ring network has a ring topology, although it looks more like a star in appearance. This is due to the fact that individual subscribers (computers) are not connected to the network directly, but through special hubs or multi-station access devices (MSAU or MAU - Multistation Access Unit). Physically, the network forms a star-ring topology (Figure 7.3). In reality, the subscribers are nevertheless united in a ring, that is, each of them transmits information to one neighboring subscriber, and receives information from another.

Fig. 7.3. Star-ring token-ring network topology

At the same time, the hub (MAU) allows you to centralize the configuration, disconnect faulty subscribers, monitor the network, etc. (fig. 7.4). It does not perform any information processing.

Fig. 7.4. Ringing Token-Ring Subscribers Using a Hub (MAU)

For each subscriber, a special Trunk Coupling Unit (TCU) is used as part of the hub, which provides automatic switching on the subscriber into the ring, if it is connected to the hub and is working properly. If a subscriber disconnects from the hub or fails, the TCU automatically restores the integrity of the ring without the participation of this subscriber. The TCU is triggered by a DC signal (the so-called phantom current) that comes from a subscriber who wants to join the ring. The subscriber can also disconnect from the ring and carry out a self-test procedure (the rightmost subscriber in Fig. 7.4). The phantom current does not affect the information signal in any way, since the signal in the ring does not have a constant component.

Constructively, the hub is autonomous unit with ten connectors on the front panel (Fig. 7.5).

Fig. 7.5. Token-Ring Hub (8228 MAU)

Eight central connectors (1 ... 8) are designed to connect subscribers (computers) using adapter cables or radial cables. The two extreme connectors: input RI (Ring In) and output RO (Ring Out) are used to connect to other hubs using special trunk cables (Path cables). Wall-mount and desktop-mount options are available.

There are both passive and active MAUs. An active hub recovers the signal coming from the subscriber (that is, it acts as an Ethernet hub). The passive hub does not perform signal recovery, only re-switches the communication lines.

The hub in the network can be the only one (as in Figure 7.4), in this case, only the subscribers connected to it are closed in the ring. Outwardly, this topology looks like a star. If more than eight subscribers need to be connected to the network, then several hubs are connected by trunk cables and form a star-ring topology.

As noted, ring topology is very sensitive to ring cable breaks. To increase the survivability of the network, Token-Ring provides a so-called ring folding mode, which allows you to bypass the break point.

In normal mode, the hubs are connected in a ring by two parallel cables, but information is transmitted only through one of them (Fig. 7.6).

Fig. 7.6. Combining MAUs in Normal Mode

In the event of a single damage (breakage) of the cable, the network transmits through both cables, thereby bypassing the damaged section. At the same time, the order of bypassing subscribers connected to concentrators is even preserved (Fig. 7.7). True, the total length of the ring increases.

In the event of multiple damage to the cable, the network splits into several parts (segments) that are not connected to each other, but remain fully operational (Fig. 7.8). The maximum part of the network remains connected, as before. Of course, this no longer rescues the network as a whole, but it allows, with the correct distribution of subscribers among the hubs, to retain a significant part of the functions of the damaged network.

Several hubs can be constructively combined into a group, a cluster, within which subscribers are also connected in a ring. The use of clusters allows you to increase the number of subscribers connected to one center, for example, up to 16 (if the cluster includes two hubs).

Fig. 7.7. Collapsing the ring when the cable is damaged

Fig. 7.8. Ring disintegration with multiple cable damage

At first, twisted pair, both unshielded (UTP) and shielded (STP), were used as a transmission medium in the IBM Token-Ring network, but then there were options for equipment for coaxial cable, as well as for fiber-optic cable in the FDDI standard.

The main technical characteristics of the classic version of the Token-Ring network:

• the maximum number of IBM 8228 MAU type hubs is 12;
• the maximum number of subscribers in the network is 96;
• maximum cable length between the subscriber and the hub - 45 meters;
• maximum cable length between hubs - 45 meters;
• the maximum length of the cable connecting all the hubs is 120 meters;
• data transfer rate - 4 Mbit / s and 16 Mbit / s.

All specifications are based on the use of an unshielded twisted pair cable. If a different transmission medium is used, the characteristics of the network may differ. For example, when using shielded twisted pair (STP), the number of subscribers can be increased to 260 (instead of 96), the cable length - up to 100 meters (instead of 45), the number of hubs - up to 33, and the total length of the ring connecting the hubs - up to 200 meters ... Fiber optic cable allows to extend the cable length up to two kilometers.

To transfer information in Token-Ring, a biphase code is used (more precisely, its version with a mandatory transition in the center of the bit interval). As with any star topology, no additional electrical termination or external grounding is required. Negotiation is performed by the hardware of the network adapters and hubs.

Token-Ring cabling uses RJ-45 (unshielded twisted pair) connectors, as well as MIC and DB9P connectors. The wires in the cable connect the same pins of the connectors (that is, the so-called straight cables are used).

The Token-Ring network in the classic version is inferior to Ethernet in both the allowable size and the maximum number of subscribers. In terms of transmission speed, there are currently 100 Mbps (High Speed ​​Token-Ring, HSTR) and 1000 Mbps (Gigabit Token-Ring) versions of Token-Ring. Companies that support Token-Ring (including IBM, Olicom, Madge) do not intend to abandon their network, seeing it as a worthy competitor to Ethernet.

Compared to Ethernet hardware, Token-Ring hardware is noticeably more expensive, since it uses a more complex method of exchange control, so the Token-Ring network is not so widespread.

However, unlike Ethernet, Token-Ring network maintains a high load level much better (more than 30-40%) and provides guaranteed access time. This is necessary, for example, in industrial networks, where a delay in reaction to an external event can lead to serious accidents.

The Token-Ring network uses the classic token access method, that is, a token constantly circulates around the ring, to which subscribers can attach their data packets (see Fig. 7.8). This implies such an important advantage of this network as the absence of conflicts, but there are also disadvantages, in particular, the need to control the integrity of the token and the dependence of the network functioning on each subscriber (in the event of a malfunction, the subscriber must be excluded from the ring).

The maximum packet transfer time in Token-Ring is 10 ms. With a maximum number of 260 subscribers, the full cycle of the ring will be 260 x 10 ms = 2.6 s. During this time, all 260 subscribers will be able to transfer their packages (if, of course, they have something to transfer). During this time, a free marker will surely reach every subscriber. This interval is also the upper limit of the Token-Ring access time.

Each subscriber of the network (its network adapter) must perform the following functions:

• identification of transmission errors;
• network configuration control (network restoration in case of failure of the subscriber that precedes him in the ring);
• control of numerous time relationships adopted in the network.

The large number of functions, of course, complicates and increases the cost of the network adapter hardware.

To control the integrity of the token in the network, one of the subscribers is used (the so-called active monitor). At the same time, its hardware is no different from the others, but its software monitors the time relationships in the network and generates a new marker if necessary.

The active monitor performs the following functions:

• launches a marker into the ring at the beginning of work and when it disappears;
• regularly (every 7 seconds) informs about his presence with a special control package (AMP - Active Monitor Present);
• removes from the ring a packet that was not deleted by the subscriber who sent it;
• monitors the allowed packet transmission time.

The active monitor is selected when the network is initialized; it can be any computer on the network, but, as a rule, it becomes the first subscriber connected to the network. The subscriber, who has become an active monitor, includes its buffer (shift register) into the network, which guarantees that the marker will fit into the ring even with the minimum ring length. The size of this buffer is 24 bits for 4 Mbps and 32 bits for 16 Mbps.

Each subscriber constantly monitors how the active monitor performs its duties. If the active monitor fails for some reason, a special mechanism is activated through which all other subscribers (spare, backup monitors) decide on the appointment of a new active monitor. To do this, the subscriber who detects the failure of the active monitor transmits a control packet (token request packet) with its MAC address over the ring. Each subsequent subscriber compares the MAC address from the packet with its own. If his own address is less, he passes the packet on unchanged. If more, then it sets its own MAC address in the packet. The active monitor will be the subscriber whose MAC address value is higher than that of the others (he must receive back the packet with his MAC address three times). A sign of failure of the active monitor is its failure to perform one of the listed functions.

The Token-Ring network token is a control packet containing only three bytes (Figure 7.9): the Start Delimiter byte (SD), Access Control byte (AC), and End Delimiter byte (ED). All these three bytes are also included in the information package, although their functions in the marker and in the package are somewhat different.

The leading and trailing delimiters are not just a sequence of zeros and ones, but contain signals of a special kind. This was done so that the delimiters could not be confused with any other packet bytes.

Fig. 7.9. Token-Ring Token Format

The initial SD delimiter contains four non-standard bit intervals (Figure 7.10). Two of them, denoted by J, represent a low signal level during the entire bit interval. The other two bits, denoted K, represent a high signal level for the entire bit interval. It is clear that such timing failures are easily detected by the receiver. Bits J and K can never appear among the bits of useful information.

Fig. 7.10. Leading (SD) and Ending (ED) Delimiter Formats

The final delimiter ED also contains four special bits (two J bits and two K bits), as well as two one bits. But, in addition, it also includes two information bits, which are meaningful only as part of an information package:

• Bit I (Intermediate) is a sign of an intermediate packet (1 corresponds to the first in a chain or intermediate packet, 0 - to the last in a chain or a single packet).
• The E (Error) bit is a sign of a detected error (0 corresponds to the absence of errors, 1 to their presence).

The Access Control (AC) byte is divided into four fields (Figure 7.11): a priority field (three bits), a marker bit, a monitor bit, and a reservation field (three bits).

Fig. 7.11. Access Control Byte Format

The priority bits (field) allow the subscriber to assign priority to his packets or token (priority can be from 0 to 7, with 7 being the highest priority and 0 being the lowest). The subscriber can attach his package to the marker only when his own priority (the priority of his packages) is the same or higher than the priority of the token.

The marker bit determines whether a packet is attached to the marker or not (one corresponds to a marker without a packet, zero - to a marker with a packet). The monitor bit, set to one, indicates that this marker was transmitted by the active monitor.

Reservation bits (field) allow the subscriber to reserve his right to further seize the network, that is, to take a queue for service. If the subscriber's priority (the priority of his packets) is higher than the current value of the reservation field, then he can write his priority there instead of the previous one. After looping around the ring, the highest priority of all subscribers will be recorded in the reservation field. The content of the reservation field is similar to the content of the priority field, but indicates the future priority.

As a result of the use of priority and reservation fields, only subscribers who have packets for transmission with the highest priority can access the network. Lower priority packets will be served only when higher priority packets are exhausted.

The format of the information packet (frame) Token-Ring is shown in Fig. 7.12. In addition to the start and end delimiters, and the access control byte, this packet also includes the packet control byte, receiver and transmitter network addresses, data, checksum, and packet status byte.

Fig. 7.12. Packet (frame) format of the Token-Ring network (the length of the fields is given in bytes)

The purpose of the fields of the packet (frame).

• The leading separator (SD) is the start of the packet, the format is the same as in the marker.
• The Access Control (AC) byte has the same format as the token.
• The Packet Control Byte (FC - Frame Control) defines the type of packet (frame).
• The six-byte source and destination MAC addresses of a packet follow the standard format described in Chapter 4.
• The data field (Data) includes the data to be transmitted (in an information packet) or information for controlling the exchange (in a control packet).
• The Frame Check Sequence (FCS) field is a 32-bit cyclic packet checksum (CRC).
• The trailing separator (ED), as in the marker, indicates the end of the packet. In addition, it determines whether the given packet is intermediate or final in the sequence of transmitted packets, and also contains a sign of packet error (see Fig. 7.10).
• The packet status byte (FS - Frame Status) tells what happened to the given packet: whether it was seen by the receiver (that is, whether there is a receiver with the specified address) and copied into the receiver's memory. From it, the sender of the packet knows whether the packet arrived at its destination and without errors, or if it needs to be transmitted again.

It should be noted that the larger allowable size of the transmitted data in one packet compared to the Ethernet network can be a decisive factor in increasing the network performance. Theoretically, for transfer rates of 16 Mbit / s and 100 Mbit / s, the length of the data field can even reach 18 Kbytes, which is essential when transferring large amounts of data. But even at 4 Mbps, Token-Ring often provides faster actual transfer rates than Ethernet (10 Mbps) due to token access. The advantage of Token-Ring is especially noticeable at high loads (over 30-40%), since in this case the CSMA / CD method requires a lot of time to resolve repeated conflicts.

A subscriber wishing to transmit a packet waits for a free token to arrive and captures it. The captured marker is transformed into the frame of the information packet. Then the subscriber transmits the information packet to the ring and waits for its return. It then releases the token and sends it back to the network.

In addition to the token and the usual packet, a special control packet can be transmitted in the Token-Ring network, which serves to interrupt the transmission (Abort). It can be sent anytime and anywhere in the data stream. This package consists of two single-byte fields - the initial (SD) and final (ED) delimiters of the described format.

Interestingly, the faster version of Token-Ring (16 Mbps and higher) uses the so-called Early Token Release (ETR) method. It avoids network overhead while the data packet is looped back to its sender.

The ETR method boils down to the fact that immediately after transmitting its packet attached to the token, any subscriber issues a new free token to the network. Other subscribers can start transmitting their packets immediately after the end of the previous subscriber's packet, without waiting for him to complete the traversal of the entire network ring. As a result, there can be several packets on the network at the same time, but there will always be no more than one free token. This pipeline is especially effective on long-haul networks that have significant propagation delay.

When a subscriber is connected to the hub, it performs the procedure of autonomous self-test and cable testing (it does not turn on in the ring yet, since there is no phantom current signal). The subscriber sends himself a number of packets and checks the correctness of their passage (his input is directly connected to his output by the TCU, as shown in Fig. 7.4). After that, the subscriber includes himself in the ring, sending a phantom current. At the moment of switching on, the packet transmitted over the ring can be corrupted. Next, the subscriber sets up synchronization and checks for an active monitor on the network. If there is no active monitor, the subscriber starts the competition for the right to become one. Then the subscriber checks the uniqueness of his own address in the ring and collects information about other subscribers. After which he becomes a full participant in the exchange over the network.

In the course of the exchange, each subscriber monitors the health of the previous subscriber (around the ring). If he suspects a failure of the previous subscriber, he starts the automatic ring recovery procedure. A special control package (buoy) tells the previous subscriber to conduct a self-test and, possibly, disconnect from the ring.

The Token-Ring network also provides for the use of bridges and switches. They are used to divide a large ring into several ring segments that can exchange packets with each other. This allows you to reduce the load on each segment and increase the proportion of time provided to each subscriber.

As a result, you can form a distributed ring, that is, the union of several ring segments into one large backbone ring (Figure 7.13) or a star-ring structure with a central switch to which the ring segments are connected (Figure 7.14).

Fig. 7.13. Connecting segments with a trunk ring using bridges

Fig. 7.14. Aggregation of segments with a central switch

Arcnet network (or ARCnet from English Attached Resource Computer Net, computer network connected resources) is one of the oldest networks. It was developed by the Datapoint Corporation back in 1977. There are no international standards for this network, although it is she who is considered the ancestor of the token access method. Despite the lack of standards, the Arcnet network until recently (in 1980 - 1990) was popular, even seriously competing with Ethernet. A large number of companies (for example, Datapoint, Standard Microsystems, Xircom, etc.) have produced equipment for this type of network. But now the production of Arcnet equipment has practically ceased.

Among the main advantages of the Arcnet network in comparison with Ethernet are the limited amount of access time, high reliability of communication, ease of diagnostics, as well as the relatively low cost of adapters. The most significant disadvantages of the network include low speed information transmission (2.5 Mbit / s), addressing system and packet format.

To transmit information in the Arcnet network, a rather rare code is used, in which a logical one corresponds to two pulses during a bit interval, and a logical zero corresponds to one pulse. Obviously this is self-timing code that requires even more cable bandwidth than even Manchester's.

A coaxial cable with a characteristic impedance of 93 Ohm, for example, of the RG-62A / U brand, is used as a transmission medium in the network. Twisted pair options (shielded and unshielded) are not widely used. Fiber optic options have been proposed, but they haven't saved Arcnet either.

As a topology, the Arcnet network uses the classic bus (Arcnet-BUS) as well as a passive star (Arcnet-STAR). Hubs are used in the star. It is possible to combine bus and star segments in a tree topology using hubs (as in Ethernet). The main limitation is that there should be no closed paths (loops) in the topology. Another limitation: the number of daisy-chained segments using hubs must not exceed three.

Hubs are of two types:

• Active concentrators (restore the shape of incoming signals and amplify them). The number of ports is from 4 to 64. Active hubs can be interconnected (cascaded).
• Passive hubs (simply mix the incoming signals without amplification). The number of ports is 4. Passive hubs cannot be connected to each other. They can only link active hubs and / or network adapters.

Bus segments can only be connected to active hubs.

There are also two types of network adapters:

• High impedance (Bus) for use in bus segments:
• Low impedance (Star) designed for use in a passive star.

Low impedance adapters differ from high impedance adapters in that they contain 93-ohm matching terminators. When using them, external approval is not required. In bus segments, low impedance adapters can be used as terminating adapters for bus termination. High impedance adapters require external 93 ohm terminators. Some network adapters have the ability to switch from a high impedance state to a low impedance state, they can work both in the bus and in the star.

Thus, the topology of the Arcnet network is as follows (Figure 7.15).

Fig. 7.15. Arcnet network topology bus type (B - adapters for working in the bus, S - adapters for working in a star)

The main technical characteristics of the Arcnet network are as follows.

• Transmission medium - coaxial cable, twisted pair.
• The maximum length of the network is 6 kilometers.
• The maximum cable length from the subscriber to the passive hub is 30 meters.
• The maximum cable length from the subscriber to the active hub is 600 meters.
• The maximum cable length between active and passive hubs is 30 meters.
• The maximum cable length between active hubs is 600 meters.
• The maximum number of subscribers in the network is 255.
• The maximum number of subscribers on a bus segment is 8.
• The minimum distance between subscribers in the bus is 1 meter.
• The maximum length of a bus segment is 300 meters.
• The data transfer rate is 2.5 Mbps.

When creating complex topologies, it is necessary to ensure that the delay in the propagation of signals in the network between subscribers does not exceed 30 μs. The maximum attenuation of the signal in the cable at a frequency of 5 MHz should not exceed 11 dB.

The Arcnet network uses a token access method (transfer of rights), but it differs slightly from that of Token-Ring. This method is closest to the one provided in the IEEE 802.4 standard. The sequence of actions of subscribers with this method:

1. The subscriber who wants to transmit waits for the token to arrive.

2. Having received the token, he sends a request to transmit information to the receiving subscriber (asks if the receiver is ready to receive his packet).

3. The receiver, having received the request, sends a response (confirms its readiness).

4. Having received confirmation of readiness, the subscriber-transmitter sends its packet.

5. On receiving the packet, the receiver sends an acknowledgment of the packet.

6. The transmitter, having received an acknowledgment of packet reception, ends its communication session. After that, the token is passed to the next subscriber in descending order of network addresses.

Thus, in this case, the packet is transmitted only when there is confidence in the readiness of the receiver to receive it. This significantly increases the reliability of the transmission.

As with Token-Ring, conflicts are completely eliminated in Arcnet. Like any token network, Arcnet holds the load well and guarantees the amount of network access time (as opposed to Ethernet). The total round trip time of all subscribers by the marker is 840 ms. Accordingly, the same interval determines the upper limit of the network access time.

The token is formed by a special subscriber - the network controller. It is the subscriber with the minimum (zero) address.

If the subscriber does not receive a free token within 840 ms, then he sends a long bit sequence to the network (to ensure the destruction of the damaged old token). After that, the procedure for monitoring the network and assigning (if necessary) a new controller is performed.

The Arcnet package size is 0.5 KB. In addition to the data field, it also includes 8-bit receiver and transmitter addresses and a 16-bit cyclic checksum (CRC). Such a small packet size turns out to be not very convenient with a high traffic intensity over the network.

Arcnet network adapters differ from other network adapters in that they need to set their own network address using switches or jumpers (there can be 255 of them, since the last, 256th address is used in the network for broadcasting mode). The control over the uniqueness of each network address is entirely the responsibility of the network users. Connecting new subscribers becomes quite difficult at the same time, since it is necessary to set the address that has not yet been used. The choice of the 8-bit address format limits the number of network subscribers to 255, which may not be enough for large companies.

As a result, all this led to the almost complete abandonment of the Arcnet network. There were 20 Mbit / s versions of the Arcnet network, but these were not widely adopted.

Articles to read:

Lecture 6: Standard Ethernet / Fast Ethernet Segments

Objectives of the work

The purpose of this work is to study the principles of Ethernet and Fast Ethernet technologies and the practical development of methods for assessing the performance of a network built on the basis of Fast Ethernet technology.

Theoretical information

Ethernet technology. The Ethernet specification was proposed by DEC, Intel and Xerox (DIX) in 1980, and a little later on it was based on the IEEE 802.3 standard.

The first versions of Ethernet vl.O and Ethernet v2.0 used only coaxial cable as the transmission medium. The IEEE 802.3 standard also allows the use of twisted pair and optical fiber as a transmission medium. In 1995, the IEEE 802.3u (Fast Ethernet) standard was adopted with a speed of 100 Mbit / s, and in 1997, IEEE 802.3z (Gigabit Ethernet - 1000 Mbit / s). In the fall of 1999 the standard IEEE 802.3ab - Gigabit Ethernet on twisted pair of category 5 was adopted.

In Ethernet notation (10BASE2, 100BASE-TX, etc.), the first element denotes the data transfer rate in Mbps; the second BASEB means that direct (unmodulated) transmission is used; the third B element denotes the rounded value of the cable length in hundreds of meters B (10BASE2 - 185 m, 10BASE5 - 500 m) or the type of transmission medium (T, TX, T2, B T4 - twisted pair; FX, FL, FB, SX and LX - fiber optic; CX - twinax cable for Gigabit Ethernet).

At the heart of Ethernet is Carrier Listening and Collision Detection Multiple Media Access - CSMA / CD

• (Carrier Sense with Multiple Access and Collision Detection), implemented by the adapters of each network node at the hardware or firmware level:
• all adapters have a medium access unit (MAU) - a transceiver connected to a common (shared) data transmission medium;
• each adapter of the node, before transmitting information, listens to the line until there is no signal (carrier);
• the adapter then generates a frame starting with a sync preamble followed by a self-sync (Manchester) binary data stream;
• other nodes receive the sent signal, synchronize with the preamble and decode it into a sequence of bits;
• the end of the frame transmission is determined by the receiver B detecting that there is no carrier;
• in case of detection collisions(collision of two signals from different nodes) the transmitting nodes stop transmitting the frame, after which, at a random time interval (each through its own), they carry out a repeated transmission attempt after the line is released; at the nextB failure, the next attempt is made (and so on up to 16 times), and the delay interval B increases;
• the collision is detected by the receiver on a non-standard frame length B, which cannot be less than 64 bytes, excluding the preamble;
• a time gap must be provided between frames ( interframe or inter-packet gap, IPG - inter-packet gap) duration B 9.6 μs - the node has no right to start transmission earlier than in the interval B IPG, after determining the moment of loss of the carrier.

Definition 1. Collision domain- a group of nodes connected by a common transmission medium (cables and repeaters).

The length of the collision domain is limited by the propagation time of the signal between the nodes most distant from each other.

Definition 2. Collision Domain Diameter- the distance between the two terminal devices farthest from each other.

Definition 3. Bit interval- the time required to transmit one bit.

The bit interval on Ethernet (at 10 Mbps) is 0.1 µs.

Fast Ethernet technology. In Fast Ethernet technology, the bit interval is 0.01 µs, which gives a tenfold increase in the data transfer rate. At the same time, the frame format, the amount of data carried by the frame and the mechanism for accessing the data transmission channel remained unchanged in comparison with Ethernet.

Fast Ethernet uses a data transmission medium for operation at a speed of 100 Mbit / s, which in the IEEE 802.3u specification is designated "100BASE-T4" and "100BASE-TX" (twisted pair); "100BASE-FX" and "100BASE-SX" (fiber optic).

Networking rules

First model of Fast Ethernet network. The model is, in fact, a set of rules for building a network (Table L.1):

• - the length of each twisted pair segment must be less than 100 m;
• - the length of each fiber optic segment must be less than 412 m;
• - if MP (Media Independent Interface) cables are used, then each of them must be less than 0.5 m;
• - delays introduced by the MP cable are not taken into account when assessing the temporal parameters of the network, since they are an integral part of the delays introduced by terminal devices (terminals) and repeaters.

Table L. 1

Maximum permissible collision domain diameter in Fast Ethernet

The standard defines two classes of repeaters:

• class I repeaters convert the input signals B to digital form, and upon transmission, re-encode the digital data B into physical signals; signal conversion in the repeater requires some time, so only one class I repeater is allowed in the collision domain;
• repeaters of class II immediately transmit the received signals without any conversion, therefore only segments can be connected to them using the same data coding methods; no more than two class II repeaters can be used in one collision domain.

The second model of the Fast Ethernet network. The second model contains a sequence of calculations of the network timing in the half-duplex mode of data exchange. The diameter of the collision domain and the number of segments in it are limited by the double turnover time required for the correct operation of the collision detection and resolution mechanism (Table L.2).

Table L2

Time delays of Fast Ethernet network components

The double turnover time is calculated for the worst (in terms of signal propagation) path between two nodes of the collision domain. The calculation is performed by summing the time delays in segments, repeaters and terminals.

To calculate the double turnover time, multiply the segment length by the specific double turnover time of the corresponding segment. After determining the round-trip times for all segments of the worst-case path, add the latency introduced by the endpoint pair and repeaters to these. To take into account unforeseen delays, it is recommended to add 4 more bit intervals (bi) V to the result obtained and compare the result with 512. If the obtained result does not exceed 512 bi, then the network is considered operational.

An example of calculating the configuration of a Fast Ethernet network. In fig. L.28 shows an example of one of the maximum permissible configurations of the Fast Ethernet network.

Fig. L.28. Example of a valid Fast Ethernet network configuration

The diameter of the collision domain is calculated as the sum of the lengths of the segments A (100 m), B (5 m) and C (100 m) and is equal to 205 m.The length of the segment connecting B repeaters can be more than 5 m, if the diameter of the collision domain does not exceed the allowed limit for this configuration. The switch (switching hub), which is part of the network (see Fig. L.28), is considered a terminal device, since collisions do not propagate through it. Therefore, a 2-kilometer segment of fiber-optic cable connecting this switch with a router (router), is not taken into account when calculating the diameter of the collision domain of the Fast Ethernet network. The network satisfies the rules of the first model.

Let us now check it using the second model. The worst paths in the collision domain are from DTE1 to DTE2 and from DTE1 to the switching hub. Both paths consist of three twisted-pair segments connected by two class II repeaters. Two segments have a maximum permissible length of 100 m. The length of the segment connecting the repeaters is 5 m.

Suppose that all three segments under consideration are 100BASE-TX segments and use Category 5 twisted pair cable. LZ shows the values ​​of the double turnover time for the paths consideredB (see Fig. L.28). Adding the numbers from the second column of this table, we get 511.96 bi - this will be the double turnover time for the worst path.

Table L.Z

Network double turnover time Fast Ethernet

It should be noted that in this case there is no safety stock of 4 bi, since in this example the worst values ​​of delays B are used (see Table L.2). Actual timing of FastB Ethernet components may vary for the better.

Task to complete

It is required to evaluate the performance of a 100-megabit Fast Ethernet network in accordance with the first and second models. Network configurations are shown in table. L.4. The network topology is shown in Fig. L. 29-L.ZO.

Table L.4

Job options

	Segment 1	Segment 2	Segment 3	Segment 4	Segment 5	Segment 6
	100BASETX, 100 m	100BASETX, 95 m	100BASETX, 80 m		100BASETX, 100 m	100BASETX, 100 m

	Segment 1	Segment 2	Segment 3	Segment 4	Segment 5	Segment 6
	YUOVABE-TX, 15 m	YUOVABE-TX, 5 m	YUOVAEE-TX, 5 m	100V ABE-EX, 400 m	YUOVABE-TX, 10 m	YUOVABE-TX, 4 m
	YUOVABE-TX, 60 m	YUOVABE-TX, 95 m	YUOVABE-TX, 10 m	YUOVABE-TX, 10 m	YUOVABE-TX, 90 m	YUOVABE-TX, 95 m

Fig. L.29. Network topology 1

Fig. L. 30. Network topology 2

Fast Ethernet

Fast Ethernet - IEEE 802.3 u specification officially adopted on October 26, 1995 defines the protocol standard link layer for networks operating using both copper and fiber-optic cables at a speed of 100 Mb / s. The new specification is the successor to the IEEE 802.3 Ethernet standard, using the same frame format, CSMA / CD media access mechanism, and star topology. The evolution has touched several elements of the configuration of the physical layer facilities, which made it possible to increase throughput, including the types of cable used, the length of the segments, and the number of hubs.

Fast Ethernet structure

To better understand the operation and understand the interaction of Fast Ethernet elements, refer to Figure 1.

Figure 1. Fast Ethernet System

Logic Link Control (LLC) Sublayer

The IEEE 802.3 u specification breaks down link layer functions into two sublayers: logical link control (LLC) and medium access layer (MAC), which will be discussed below. LLC, whose functions are defined by the IEEE 802.2 standard, actually provides interconnection with higher-level protocols (for example, IP or IPX), providing various communication services:

• Service without connection establishment and acknowledgment of receipt. A simple service that does not provide flow control or error control, and does not guarantee correct data delivery.
• Connection-oriented service. An absolutely reliable service that guarantees correct data delivery by establishing a connection to the receiving system before the data transfer begins and using error control and data flow control mechanisms.
• Connectionless service with acknowledgments. A moderately complex service that uses acknowledgment messages to ensure delivery, but does not establish connections until data is sent.

On the transmitting system, the downstream data from the Network Layer protocol is first encapsulated by the LLC sublayer. The standard calls them Protocol Data Unit (PDU). When the PDU is handed down to the MAC sublayer, where it is again framed with a header and post information, it can technically be called a frame at this point. For an Ethernet packet, this means that the 802.3 frame contains a three-byte LLC header in addition to the Network Layer data. Thus, the maximum allowable data length in each packet is reduced from 1500 to 1497 bytes.

The LLC header consists of three fields:

In some cases, LLC frames play a minor role in the network communication process. For example, on a network using TCP / IP along with other protocols, the only function of LLC might be to allow 802.3 frames to contain a SNAP header, like an Ethertype, indicating the Network Layer protocol to which the frame should be sent. In this case, all LLC PDUs use the unnumbered information format. However, other higher-level protocols require a more advanced service from the LLC. For example, NetBIOS sessions and several NetWare protocols use LLC connection-oriented services more broadly.

SNAP header

The receiving system needs to determine which of the Network Layer protocols should receive the incoming data. 802.3 packets within the LLC PDU use another protocol called Sub -NetworkAccessProtocol (SNAP, Subnetting Access Protocol).

The SNAP header is 5 bytes long and is located immediately after the LLC header in the data field of the 802.3 frame, as shown in the figure. The header contains two fields.

Organization code. The Organization or Vendor ID is a 3-byte field that takes the same value as the first 3 bytes of the sender's MAC address in the 802.3 header.

Local code. The local code is a 2 byte field that is functionally equivalent to the Ethertype field in the Ethernet II header.

Matching sublevel

As stated earlier, Fast Ethernet is an evolved standard. A MAC designed for the AUI interface needs to be mapped for the MII interface used in Fast Ethernet, which is what this sublayer is for.

Media Access Control (MAC)

Each node in a Fast Ethernet network has a media access controller (MediaAccessController- MAC). MAC is key to Fast Ethernet and has three purposes:

The most important of the three MAC assignments is the first. For any networking technology that uses a common medium, the medium access rules that determine when a node can transmit are its primary characteristic. Several IEEE committees are involved in the development of rules for accessing the environment. The 802.3 committee, often referred to as the Ethernet committee, defines LAN standards that use rules called CSMA /CD(Carrier Sense Multiple Access with Collision Detection).

CSMS / CD are media access rules for both Ethernet and Fast Ethernet. It is in this area that the two technologies completely coincide.

Since all nodes in Fast Ethernet share the same medium, they can only transmit when it is their turn. This queue is defined by CSMA / CD rules.

CSMA / CD

The MAC Fast Ethernet controller listens on the carrier before transmitting. The carrier exists only when another node is transmitting. The PHY layer detects the presence of a carrier and generates a message for the MAC. The presence of a carrier indicates that the medium is busy and the listening node (or nodes) must yield to the transmitting one.

A MAC that has a frame to transmit must wait a minimum amount of time after the end of the previous frame before transmitting it. This time is called interpacket gap(IPG, interpacket gap) and lasts 0.96 microseconds, that is, a tenth of the transmission time of a regular Ethernet packet at 10 Mbps (IPG is the only time interval, always specified in microseconds, not bit time) Figure 2.

Figure 2. Interpacket gap

After the end of packet 1, all LAN nodes must wait for the IPG time before being able to transmit. The time interval between packets 1 and 2, 2 and 3 in Fig. 2 is the IPG time. After the transmission of packet 3 was complete, no nodes had material to process, so the time interval between packets 3 and 4 is longer than the IPG.

All nodes on the network must comply with these rules. Even if a node has many frames to transmit and this node is the only transmitting one, then after sending each packet it must wait for at least IPG time.

This is part of the CSMA Fast Ethernet Media Access Rules. In short, many nodes have access to the medium and use the carrier to keep track of whether it is busy.

The early experimental networks applied exactly these rules, and such networks worked very well. However, the use of CSMA alone led to a problem. Often, two nodes, having a packet to transmit and waiting for IPG time, would start transmitting at the same time, resulting in data corruption on both sides. This situation is called collision(collision) or conflict.

To overcome this obstacle, early protocols used a fairly simple mechanism. Packages were divided into two categories: commands and reactions. Each command sent by the node required a response. If no response was received for some time (called a timeout period) after the command was sent, the original command was re-issued. This could happen several times (the maximum number of timeouts) before the sending node recorded the error.

This scheme could work fine, but only up to a certain point. The occurrence of conflicts led to a sharp decrease in performance (usually measured in bytes per second), because nodes often stood idle waiting for responses to commands that never reached their destination. Network congestion, an increase in the number of nodes are directly related to an increase in the number of conflicts and, consequently, to a decrease in network performance.

Early network designers quickly found a solution to this problem: each node must detect the loss of a transmitted packet by detecting a conflict (and not wait for a reaction that will never follow). This means that packets lost due to the conflict must be re-transmitted immediately before the timeout expires. If the host transmitted the last bit of the packet without a conflict, then the packet was transmitted successfully.

Carrier sense can be combined well with collision detection. Collisions still continue to occur, but this does not affect the performance of the network, since the nodes quickly get rid of them. The DIX group, having developed the rules for accessing the CSMA / CD environment for Ethernet, formalized them in the form of a simple algorithm - Figure 3.

Figure 3. Algorithm of CSMA / CD operation

Physical layer device (PHY)

Since Fast Ethernet can use different type cable, a unique signal preconversion is required for each medium. Conversion is also required for efficient data transmission: to make the transmitted code resistant to interference, possible loss, or distortion of its individual elements (baud), to ensure effective synchronization of clocks on the transmitting or receiving side.

Coding Sub-Layer (PCS)

Encodes / decodes data coming from / to the MAC layer using algorithms or.

Physical interconnection and physical media dependency sublayers (PMA and PMD)

The PMA and PMD sublayers communicate between the PSC sublayer and the MDI interface, providing formation in accordance with the physical coding method: or.

Auto-negotiation sublevel (AUTONEG)

The auto-negotiation sublayer allows two communicating ports to automatically select the most efficient mode of operation: full-duplex or half-duplex 10 or 100 Mb / s. Physical layer

The Fast Ethernet standard defines three types of 100 Mbps Ethernet signaling media.

• 100Base-TX - two twisted pairs of wires. Transmission is carried out in accordance with the standard for data transmission in a twisted physical medium, developed by ANSI (American National Standards Institute - American National Standards Institute). Coiled data cable can be shielded or unshielded. Uses 4B / 5B data coding algorithm and MLT-3 physical coding method.
• 100Base-FX is a two-core fiber optic cable. Transmission is also carried out in accordance with the ANSI standard for data transmission in fiber optic media. Uses 4B / 5B data coding algorithm and NRZI physical coding method.

100Base-TX and 100Base-FX specifications are also known as 100Base-X

• 100Base-T4 is a special specification developed by the IEEE 802.3u committee. According to this specification, data transmission is carried out over four twisted pairs of telephone cable, which is called UTP Category 3 cable. It uses 8B / 6T data coding algorithm and NRZI physical coding method.

Additionally, the Fast Ethernet standard includes guidelines for Category 1 shielded twisted pair cable, which is the standard cable traditionally used in Token Ring networks. The support organization and guidelines for using STP cable on Fast Ethernet provide a fast Ethernet migration path for customers with STP cabling.

The Fast Ethernet specification also includes an auto-negotiation mechanism that allows a host port to automatically adjust to a data transfer rate of 10 Mbps or 100 Mbps. This mechanism is based on the exchange of a number of packets with a port of a hub or switch.

100Base-TX environment

Two twisted pairs are used as the transmission medium for 100Base-TX, with one pair being used to transmit data and the other to receive them. Since the ANSI TP-PMD specification contains descriptions of both shielded and unshielded twisted pairs, the 100Base-TX specification includes support for both unshielded and shielded type 1 and 7 twisted pairs.

MDI (Medium Dependent Interface) connector

The media-dependent 100Base-TX link interface can be one of two types. For unshielded twisted-pair cable, use an 8-pin RJ 45 Category 5 connector as the MDI connector. The same connector is used on a 10Base-T network, which provides backward compatibility with existing Category 5 cabling. For shielded twisted pairs, an MDI connector must be used as an MDI connector. use IBM STP type 1 connector, which is a shielded DB9 connector. This connector is commonly used in Token Ring networks.

Category 5 (e) UTP cable

The UTP 100Base-TX media interface uses two pairs of wires. To minimize crosstalk and possible signal distortion, the remaining four wires should not be used to carry any signals. The transmit and receive signals for each pair are polarized, with one wire carrying a positive (+) signal and the other a negative (-) signal. The color coding of the cable wires and the pin numbers of the connector for the 100Base-TX network are shown in table. 1. Although the 100Base-TX PHY layer was developed after the adoption of the ANSI TP-PMD standard, the RJ 45 connector pin numbers have been changed to align with the 10Base-T pinouts already used. The ANSI TP-PMD standard uses pins 7 and 9 to receive data, while the 100Base-TX and 10Base-T standards use pins 3 and 6 for this. This wiring allows you to use 100Base-TX adapters instead of 10 Base adapters - T and connect them to the same Category 5 cables without changing the wiring. In the RJ 45 connector, the pairs of wires used are connected to pins 1, 2 and 3, 6. For the correct connection of the wires, follow their color coding.

Table 1. Purpose of connector contactsMDIcableUTP100Base-TX

Nodes interact with each other by exchanging frames (frames). In Fast Ethernet, a frame is the basic unit of exchange over a network - any information transmitted between nodes is placed in the data field of one or more frames. Forwarding frames from one node to another is possible only if there is a way to unambiguously identify all network nodes. Therefore, each node on a LAN has an address called its MAC address. This address is unique: no two LAN nodes can have the same MAC address. Moreover, in no LAN technology (with the exception of ARCNet) no two nodes in the world can have the same MAC address. Any frame contains at least three main pieces of information: recipient address, sender address, and data. Some frames have other fields, but only the three listed are required. Figure 4 shows the Fast Ethernet frame structure.

Figure 4. Frame structureFastEthernet

• address of the recipient- the address of the node receiving the data is indicated;
• sender's address- the address of the node that sent the data is indicated;
• length / type(L / T - Length / Type) - contains information about the type of transmitted data;
• frame checksum(PCS - Frame Check Sequence) - designed to check the correctness of the frame received by the receiving node.

The minimum frame size is 64 octets, or 512 bits (terms octet and byte - synonyms). The maximum frame size is 1518 octets, or 12144 bits.

Frame addressing

Each node on a Fast Ethernet network has a unique number called the MAC address or node address. This number consists of 48 bits (6 bytes), assigned to the network interface during device manufacture and programmed during initialization. Therefore, the network interfaces of all LANs, with the exception of ARCNet, which uses 8-bit addresses assigned by the network administrator, have a built-in unique MAC address that differs from all other MAC addresses on Earth and is assigned by the manufacturer in agreement with the IEEE.

To facilitate the management of network interfaces, the IEEE has proposed to divide the 48-bit address field into four parts, as shown in Figure 5. The first two bits of the address (bits 0 and 1) are address type flags. The meaning of the flags determines how the address part is interpreted (bits 2 - 47).

Figure 5. Format of the MAC address

The I / G bit is called individual / group address flag and shows what (individual or group) the address is. An individual address is assigned to only one interface (or node) on the network. Addresses with the I / G bit set to 0 are MAC addresses or node addresses. If the I / O bit is set to 1, then the address belongs to the group and is usually called multipoint address(multicast address) or functional address(functional address). A multicast address can be assigned to one or more LAN network interfaces. Frames sent to a multicast address receive or copy all LAN network interfaces that have it. Multicast addresses allow a frame to be sent to a subset of hosts on a local network. If the I / O bit is set to 1, then bits 46 to 0 are treated as a multicast address and not as the U / L, OUI, and OUA fields of the normal address. The U / L bit is called universal / local control flag and determines how the address was assigned to the network interface. If both bits, I / O and U / L, are set to 0, then the address is the unique 48-bit identifier described earlier.

OUI (organizationally unique identifier - organizationally unique identifier). The IEEE assigns one or more OUIs to each manufacturer of network adapters and interfaces. Each manufacturer is responsible for the correct assignment of the OUA (organizationally unique address - organizationally unique address), which should have any device it creates.

When the U / L bit is set, the address is locally managed. This means that it is not specified by the manufacturer of the network interface. Any organization can create its own MAC address for the network interface by setting the U / L bit to 1, and bits 2 through 47 to some chosen value. The network interface, having received the frame, first of all decodes the destination address. When the I / O bit is set in the address, the MAC layer will receive this frame only if the destination address is in the list that is stored on the node. This technique allows one node to send a frame to many nodes.

There is a special multicast address called broadcast address. In a 48-bit IEEE broadcast address, all bits are set to 1. If a frame is transmitted with a destination broadcast address, then all nodes on the network will receive and process it.

Field Length / Type

The L / T (Length / Type) field serves two different purposes:

• to determine the length of the data field of the frame, excluding any padding with spaces;
• to denote the data type in the data field.

The L / T field value between 0 and 1500 is the length of the data field of the frame; a higher value indicates the type of protocol.

In general, the L / T field is a historical residue of the Ethernet standardization in the IEEE, which gave rise to a number of compatibility problems for equipment released before 1983. Nowadays Ethernet and Fast Ethernet never use L / T fields. The specified field serves only for coordination with the software that processes frames (that is, with protocols). But the only truly standard purpose of the L / T field is to use it as a length field - the 802.3 specification does not even mention its possible use as a data type field. The standard states: "Frames with a length field value greater than that specified in clause 4.4.2 may be ignored, discarded, or privately used. The use of these frames is outside the scope of this standard."

Summarizing what has been said, we note that the L / T field is the primary mechanism by which frame type. Fast Ethernet and Ethernet frames, in which the length of the L / T field is set (the L / T value 802.3, frames in which the data type is set by the value of the same field (L / T value> 1500) are called frames Ethernet- II or DIX.

Data field

In the data field contains information that one node sends to another. Unlike other fields that store very specific information, a data field can contain almost any information, as long as its size is at least 46 and no more than 1500 bytes. How the content of a data field is formatted and interpreted is determined by the protocols.

If it is necessary to send data less than 46 bytes in length, the LLC layer adds bytes with an unknown value to the end of the data, called insignificant data(pad data). As a result, the field length becomes 46 bytes.

If the frame is of 802.3 type, the L / T field indicates the amount of valid data. For example, if a 12-byte message is being sent, then the L / T field contains the value 12, and the data field contains 34 additional insignificant bytes. Adding insignificant bytes initiates the Fast Ethernet LLC layer, and is usually implemented in hardware.

The MAC layer facility does not specify the contents of the L / T field — the software does. Setting the value of this field is almost always done by the network interface driver.

Frame checksum

The Frame Check Sequence (PCS) ensures that the received frames are not corrupted. When forming the transmitted frame at the MAC level, a special mathematical formula is used CRC(Cyclic Redundancy Check), designed to calculate a 32-bit value. The resulting value is placed in the FCS field of the frame. The values ​​of all bytes of the frame are supplied to the input of the MAC layer element that calculates the CRC. The FCS field is the primary and most important Fast Ethernet error detection and correction mechanism. Starting from the first byte of the destination address and ending with the last byte of the data field.

DSAP and SSAP Field Values

DSAP / SSAP Values			Description
			Indiv LLC Sublayer Mgt
			Group LLC Sublayer Mgt
			SNA Path Control
			Reserved (DOD IP)
			ISO CLNS IS 8473
			The 8B6T coding algorithm converts an eight-bit data octet (8B) to a six-bit ternary symbol (6T). Code groups 6T are designed to be transmitted in parallel over three twisted pairs of cable, so the effective data rate for each twisted pair is one third of 100 Mbit / s, that is, 33.33 Mbit / s. The ternary symbol rate for each twisted pair is 6/8 of 33.3 Mbps, which corresponds to a clock rate of 25 MHz. It is with this frequency that the timer of the MP interface works. Unlike binary signals, which have two levels, ternary signals transmitted on each pair can have three levels. Character encoding table Linear code Symbol MLT-3 Multi Level Transmission - 3 (multilevel transmission) - a bit similar to the NRZ code, but unlike the latter, it has three signal levels. The unit corresponds to the transition from one signal level to another, and the change in the signal level occurs sequentially taking into account the previous transition. When transmitting "zero", the signal does not change. This code, like NRZ, needs to be pre-encoded. Compiled on the basis of materials: Laem Queen, Richard Russell "Fast Ethernet"; K. Zakler "Computer Networks"; V.G. and N.A. Olifer "Computer Networks";

The ComputerPress test laboratory tested Fast Ethernet network cards designed for use in 10/100 Mbit / s workstations. PCI bus... The most common currently used cards with a bandwidth of 10/100 Mbit / s were selected, since, firstly, they can be used in Ethernet, Fast Ethernet and mixed networks, and, secondly, the promising Gigabit Ethernet technology ( throughput up to 1000 Mbit / s) is still used most often to connect powerful servers to the network equipment of the network core. It is extremely important what quality passive network equipment (cables, sockets, etc.) is used on the network. It is well known that while Category 3 twisted pair cable is sufficient for Ethernet networks, Category 5 is required for Fast Ethernet. Signal scattering, poor noise immunity can significantly reduce network bandwidth.

The purpose of testing was to determine, first of all, the index of effective performance (Performance / Efficiency Index Ratio - hereinafter P / E-index), and only then - the absolute value of the throughput. The P / E index is calculated as the ratio of the bandwidth of the network card in Mbps to the percentage of the CPU utilization. This index is the industry standard for determining the performance of network adapters. It was introduced in order to take into account the use by network cards of CPU resources. This is because some network adapter manufacturers try to maximize performance by using more CPU cycles on the computer to perform network operations. Low CPU usage and relatively high bandwidth are essential for running mission-critical business and multimedia applications, as well as real-time tasks.

We have tested the cards that are currently most often used for workstations in corporate and local networks:

1. D-Link DFE-538TX
2. SMC EtherPower II 10/100 9432TX / MP
3. 3Com Fast EtherLink XL 3C905B-TX-NM
4. Compex RL 100ATX
5. Intel EtherExpress PRO / 100 + Management
6. CNet PRO-120
7. NetGear FA 310TX
8. Allied Telesyn AT 2500TX
9. Surecom EP-320X-R

The main characteristics of the tested network adapters are shown in Table. one . Let us explain some of the terms used in the table. Automatic detection of the connection speed means that the adapter itself determines the maximum possible speed of operation. In addition, if autosensing is supported, no additional configuration is required when switching from Ethernet to Fast Ethernet and vice versa. That is from system administrator no need to reconfigure the adapter and reload drivers.

Bus Master support allows data transfer directly between the network card and the computer memory. This frees up the central processor to perform other operations. This property has become the de facto standard. No wonder all known network cards support the Bus Master mode.

Remote wake-on (Wake on LAN) allows you to turn on the PC over the network. That is, it becomes possible to service the PC outside of working hours. For this purpose, three-pin connectors are used on motherboard and a network adapter, which are connected with a special cable (included in the delivery set). In addition, special control software is required. Wake on LAN technology is developed by the Intel-IBM alliance.

Full duplex mode allows data to be transmitted simultaneously in both directions, half duplex - in only one. Thus, the maximum possible throughput in full duplex mode is 200 Mbps.

DMI (Desktop Management Interface) provides the ability to obtain information about the configuration and resources of the PC using network management software.

Support for the WfM (Wired for Management) specification enables the network adapter to interact with network management and administration software.

To remotely boot a computer's OS over a network, network adapters are supplied with a special BootROM memory. This allows for efficient use of diskless workstations on the network. Most tested cards only had a BootROM slot; the BootROM itself is usually a separately ordered option.

ACPI (Advanced Configuration Power Interface) support helps reduce power consumption. ACPI is a new technology for power management. It is based on the use of both hardware and software tools... Basically, Wake on LAN is an integral part of ACPI.

Proprietary means of increasing productivity can increase the efficiency of the network card. The most famous of them are Parallel Tasking II by 3Com and Adaptive. Technology company Intel. These funds are usually patented.

Support for major operating systems is provided by almost all adapters. The main operating systems include: Windows, Windows NT, NetWare, Linux, SCO UNIX, LAN Manager and others.

The level of service support is assessed by the availability of documentation, a diskette with drivers and the ability to download the latest drivers from the company's website. Packaging also plays an important role. From this point of view, the best, in our opinion, are D-Link, Allied Telesyn and Surecom network adapters. But in general, the level of support was satisfactory for all cards.

Typically, the warranty covers the entire life of the power adapter (lifetime warranty). Sometimes it is limited to 1-3 years.

Testing methodology

All tests used the latest NIC drivers downloaded from the respective vendors' Internet servers. In the case when the driver of the network card allowed any adjustments and optimizations, the default settings were used (except for the Intel network adapter). Note that the cards and corresponding drivers from 3Com and Intel have the richest additional features and functions.

Performance was measured using Novell's Perform3 utility. The principle of operation of the utility is that a small file is copied from a workstation to a shared one. network drive server, after which it remains in the file cache of the server and is read from there repeatedly within a specified period of time. This allows for memory-to-memory-to-memory interactions and eliminates the impact of disk latency. The utility parameters include initial file size, final file size, resizing step, and test time. Novell Perform3 utility displays performance values ​​with different file sizes, average and maximum performance (in KB / s). The following parameters were used to configure the utility:

• Initial file size - 4095 bytes
• Final file size - 65,535 bytes
• File increment - 8192 bytes

The test time with each file was set to twenty seconds.

Each experiment used a pair of identical network cards, one running on a server and the other running on a workstation. This does not seem to be in line with common practice, as servers usually use specialized network adapters with a number of additional features. But this is exactly how - the same network cards are installed both on the server and on workstations - testing is carried out by all well-known test laboratories in the world (KeyLabs, Tolly Group, etc.). The results are somewhat lower, but the experiment turns out to be clean, since only the analyzed network cards work on all computers.

Compaq DeskPro EN client configuration:

• Pentium II 450 MHz processor
• cache 512 KB
• RAM 128 MB
• hard drive 10 GB
• operating system Microsoft Windows NT Server 4.0 c 6 a SP
• TCP / IP protocol.

Compaq DeskPro EP server configuration:

• Celeron 400 MHz processor
• RAM 64 MB
• hard drive 4,3 GB
• operating system Microsoft Windows NT Workstation 4.0 c c 6 a SP
• TCP / IP protocol.

Testing was conducted under conditions where computers were directly connected with a UTP Category 5 crossover cable. During these tests, the cards were operating in 100Base-TX Full Duplex mode. In this mode, the throughput turns out to be slightly higher due to the fact that part of the service information (for example, acknowledgment of receipt) is transmitted simultaneously with useful information, the volume of which is estimated. In these conditions, it was possible to fix rather high values ​​of the throughput; for example, 3Com Fast EtherLink XL 3C905B-TX-NM adapter averages 79.23 Mbps.

The processor load was measured on the server using Windows utilities NT Performance Monitor; the data was written to a log file. The Perform3 utility was run on the client so as not to affect the server processor load. Intel Celeron was used as the processor of the server computer, the performance of which is significantly lower than the performance of Pentium II and III processors. Intel Celeron was used deliberately: the fact is that, since the processor load is determined with a sufficiently large absolute error, in the case of large absolute values, the relative error turns out to be smaller.

After each test, Perform3 utility places the results of its work in a text file as a dataset of the following form:

65535 bytes. 10491.49 KBps. 10491.49 Aggregate KBps. 57343 bytes. 10844.03 KBps. 10844.03 Aggregate KBps. 49151 bytes. 10737.95 KBps. 10737.95 Aggregate KBps. 40959 bytes. 10603.04 KBps. 10603.04 Aggregate KBps. 32767 bytes. 10497.73 KBps. 10497.73 Aggregate KBps. 24575 bytes. 10220.29 KBps. 10220.29 Aggregate KBps. 16383 bytes. 9573.00 KBps. 9573.00 Aggregate KBps. 8191 bytes. 8195.50 KBps. 8195.50 Aggregate KBps. 10844.03 Maximum KBps. 10145.38 Average KBp.

The file size is displayed, the corresponding throughput for the selected client and for all clients (in this case, there is only one client), as well as the maximum and average throughput throughout the test. The resulting average values ​​for each test were converted from KB / s to Mbit / s using the formula:
(KB x 8) / 1024,
and the value of the P / E index was calculated as the ratio of the throughput to the processor load in percent. Subsequently, the average value of the P / E index was calculated based on the results of three measurements.

Using the Perform3 utility on Windows NT Workstation, the following problem arose: in addition to writing to a network drive, the file was also written to the local file cache, from which it was subsequently read very quickly. The results were impressive, but unrealistic, as there was no data transfer per se over the network. To enable applications to treat shared network drives as normal local drives, in operating system a special network component is used - a redirector that redirects I / O requests over the network. Under normal operating conditions, when executing the procedure for writing a file to a shared network drive, the redirector uses the Windows NT caching algorithm. That is why, when writing to the server, it also writes to the local file cache of the client machine. And for testing, it is necessary that caching is carried out only on the server. To prevent caching on the client computer, the values ​​of the parameters in the Windows NT registry were changed, which made it possible to disable the caching performed by the redirector. Here's how it was done:

1. Registry path:

   HKEY_LOCAL_MACHINE \ SYSTEM \ CurrentControlSet \ Services \ Rdr \ Parameters

   Parameter name:

   UseWriteBehind enables write-behind optimization for files being written

   Type: REG_DWORD

   Value: 0 (default: 1)

2. Registry path:

   HKEY_LOCAL_MACHINE \ SYSTEM \ CurrentControlSet \ Services \ Lanmanworkstation \ parameters

   Parameter name:

   UtilizeNTCaching specifies whether the redirector will use the Windows NT cache manager to cache file contents.

   Type: REG_DWORD Value: 0 (default: 1)

Intel EtherExpress PRO / 100 + Management Network Adapter

The card's throughput and processor utilization are nearly the same as that of 3Com. The windows for setting parameters of this map are shown below

The new Intel 82559 controller in this card provides very high performance, especially in Fast Ethernet networks.

The technology that Intel uses in its Intel EtherExpress PRO / 100 + card is called Adaptive Technology. The essence of the method is to automatically change the time intervals between Ethernet packets, depending on the network load. As network congestion increases, the distance between individual Ethernet packets dynamically increases, which reduces collisions and increases throughput. With a low network load, when the probability of collisions is low, the time intervals between packets are reduced, which also leads to increased performance. The benefits of this method should be greatest in large collisional Ethernet segments, that is, in cases where hubs rather than switches dominate the network topology.

New Intel technology, called Priority Packet, allows you to regulate the traffic passing through the network card according to the priorities of individual packets. This provides the ability to increase data transfer rates for mission-critical applications.

VLAN support is provided (IEEE 802.1Q standard).

There are only two indicators on the board - work / connection, speed 100.

www.intel.com

SMC EtherPower II 10/100 SMC9432TX / MP Network Adapter

The architecture of this card uses two promising technologies SMC SimulTasking and Programmable InterPacket Gap. The first technology is similar to 3Com Parallel Tasking technology. Comparing the test results for cards from these two manufacturers, we can conclude about the degree of efficiency of these technologies implementation. Note also that this network card showed the third result in terms of performance and P / E index, outperforming all cards except 3Com and Intel.

There are four LED indicators on the card: speed 100, transmission, connection, duplex.

The company's main Web site is www.smc.com