Built-in ethernet. Distinctive Features of Ethernet Technology

Ethernet is the most widely used local area network standard today. Total number of networks currently using

fast ethernet

Fast Ethernet technology is similar to traditional Ethernet technology in many ways, but it is 10 times faster. Fast Ethernet or 100BASE-T operates at 100 megabits per second (Mbps) instead of 10 for the traditional Ethernet option. 100BASE-T technology uses frames of the same format and length as Ethernet and does not require changes in upper layer protocols, applications, or network operating systems on workstations. You can route and switch packets between 10 Mbps and 100 Mbps networks without protocol translation and associated delays. Fast Ethernet technology uses the CSMA/CD MAC protocol to provide media access. Most modern Ethernet networks are built on a star topology, where the hub is the center of the network, and cables from the hub run to each computer. The same topology is used in Fast Ethernet networks, although the network diameter is somewhat smaller due to the higher speed. Fast Ethernet uses unshielded twisted-pair (UTP) cable as specified in the IEEE 802.3u specification for 100BASE-T. The standard recommends the use of category 5 cable with two or four pairs of plastic-sheathed conductors. Category 5 cables are certified for 100 MHz bandwidth. In 100BASE-TX, one pair is used for data transmission, the other pair for collision detection and reception.

The Fast Ethernet standard defines three modifications for working with different types of cables: 100Base TX, 100Base T4 and 100Base FX. Modifications 100Base TX and 100Base T4 are designed for twisted pair, and 100Base FX was designed for optical cable.

The 100Base TX standard requires two shielded or unshielded twisted pairs. One pair is for transmitting, the other for receiving. Two major cabling standards meet these requirements: Category 5 Unshielded Twisted Pair (UTP-5) and IBM Type 1 Shielded Twisted Pair.

The 100Base T4 standard has less restrictive cable requirements, as it uses all four pairs of an eight-core cable: one pair for transmit, one for receive, and the remaining two pairs work both for transmission and reception. As a result, in the 100Base T4 standard, data can be received and transmitted over three pairs. For 100Base T4 networks, Category 3-5 unshielded twisted pair and Type 1 shielded twisted pair cables are suitable.

The succession of Fast Ethernet and Ethernet technologies makes it easy to develop recommendations for use: Fast Ethernet is useful in those organizations that have widely used classic Ethernet, but today are in need of increased bandwidth. At the same time, all the accumulated experience with Ethernet and, in part, the network infrastructure is preserved.

For classic Ethernet, the network listening time is determined by the maximum distance that a 512-bit frame can travel over the network in a time equal to the processing time of this frame at the workstation. For an Ethernet network, this distance is 2500 meters. In a Fast Ethernet network, the same 512-bit frame will travel only 250 meters in the time it takes to process it at a workstation.

Fast Ethernet's main area of ​​work today is workgroup and departmental networks. It is advisable to make the transition to Fast Ethernet gradually, leaving Ethernet where it does its job well. One obvious case where Ethernet should not be replaced by Fast Ethernet technology is when old personal computers with an ISA bus are connected to the network.

Gigabit Ethernet/

this technology uses the same frame format, the same CSMA/CD media access method, the same flow control mechanisms and the same control objects, yet Gigabit Ethernet is more different from Fast Ethernet than Fast Ethernet is from Ethernet. In particular, if Ethernet was characterized by a variety of supported transmission media, which gave reason to say that it could even work over barbed wire, then in Gigabit Ethernet, fiber optic cables become the dominant transmission medium (this, of course, is far from the only difference , but the rest will be discussed in more detail below). In addition, Gigabit Ethernet poses incomparably more complex technical challenges and places much higher demands on the quality of the wiring. In other words, it is much less versatile than its predecessors.

GIGABIT ETHERNET STANDARDS

The main effort of the IEEE 802.3z Working Group is to define the physical standards for Gigabit Ethernet. She took the ANSI X3T11 Fiber Channel standard as a basis, more precisely, its two lower sublevels: FC-0 (interface and transmission medium) and FC-1 (encoding and decoding). The physical media-specific specification for Fiber Channel currently specifies 1.062 Gigabauds per second. In Gigabit Ethernet, this has been increased to 1.25 gigabytes per second. Considering the 8B/10B encoding, we get a data transfer rate of 1 Gbps.

Considering the principles of operation of any technology, it is worth starting with the history of its creation. Ethernet technology came about as one of many projects at Xerox PARC Corporation. In 1973, Robert Metcalfe, an employee of the Xerox research center, wrote a memorandum describing the principles of Ethernet technology. Ethernet technology was based on the Carrier Sense Multiple Access with Collision Detection (CSMA/CD) principle. In the same year, together with David Boggs, he created the first network that connected two computers at a speed of 2.944 Mbps.

Over the years, thanks to the efforts of Robert Metcalfe, the leading companies Intel, Xerox, DEC begin to standardize the Ethernet protocol. Soon, Ethernet technology began to compete with the leading Token Ring and Arcnet technologies at the time.

In 1985, the IEEE 802.3 document was published, which describes the standard for data transmission at a speed of 10 Mbps. The first Ethernet standards used coaxial cable as the transmission medium. That is, there were no switches familiar to us then. Special transceivers or connectors were used to connect to the computer's network card. The coaxial cable acted as a common bus. Terminators were installed at both ends of the bus - network terminations. There were two varieties of the first Ethernet: 10Base5 (thick coaxial cable) and 10Base2 (thin coaxial cable).

In 1991, the 10Base-T standard was adopted, which uses a double unshielded twisted pair as a transmission medium. Category 3 (Cat 3) cable is used. Connections of end stations were carried out according to the "point-to-point" topology with a special device - a multiport repeater (hub). The principle of operation of the concentrator is quite simple. It receives a signal on one of the ports, and then repeats it to all the others. Thus, the "common bus" topology inherent in Ethernet is implemented, with bandwidth sharing between all hosts on the network.

On October 26, 1995, the IEEE officially adopted the 802.3u standard, which describes Fast Ethrenet technology. Fast Ethernet was characterized by a high data transfer rate of 100 Mbps. The CSMA/CD random access method, frame format, and star topology have been retained from traditional Ethernet. All differences from Ethernet are concentrated at the physical layer. Three types of cables are used in a Fast Ethernet organization: optical multifiber cable (100Base-FX), category 5 twisted pair (100Base-TX), category 3 twisted pair (100Base-T4).

Over time, the requirements for data transfer speeds increase. The next step in development was the standardization of the Gigabit Ethernet standard, officially called IEEE 802.3z. This standard was published in July 1998. IEEE 802.3z included three types of cables: 1000BASE-SX for signal transmission over multimode fiber, 1000BASE-LX for single-mode fiber, and the almost obsolete 1000BASE-CX for shielded balanced copper cable.

After a brief historical overview, let's go directly to the principles of Ethernet technology. At the beginning of the article, it was mentioned that Ethernet uses a Carrier Sense Multiple Access with Collision Detection (CSMA/CD) technique. It is this principle that is the "foundation" of the entire technology. What does he represent?

All stations are connected to a common bus. Each of them listens to the medium for the presence of a carrier. The presence of a carrier means that one of the stations is currently transmitting a frame. To gain access to the transmission medium, the station must detect the absence of a carrier, wait for a technological pause, and, if there is no carrier, it can start transmitting its frame. The frame is transmitted over a common bus and reaches all stations. If the destination address matches, then the station accepts the frame, otherwise it discards it.

If the stations transmit frames at the same time, then there will be collision.

Collision - superposition of two or more frames

After detecting a collision, all stations are required to stop transmitting frames and wait for a short random amount of time in order to gain access to the transmission medium again.

It can be seen from the description of the method that it is probabilistic in nature. It is assumed that any station at any time can start transmitting frames. With the increase in stations, the probability of collisions increases, as a result of which the Ethrenet standard sets a limit of no more than 1024 nodes in one network. In this case, the maximum distance between any two nodes should be no more than 2500 m.

The 802.3 standard defines the Ethrenet frame format.

Consider the frame format:

• Preamble - is a sequence of bits 10101010 ..., consisting of 7 bytes. The preamble is designed to synchronize the transceivers.
• SA (Start Delimiter) - the initial delimiter. Consists of one byte and is the sequence 10101011. This combination indicates the beginning of the frame.
• Destination address - destination address. Consists of 6 bytes and indicates the MAC address of the recipient.
• Source address - source address. Indicates the MAC address of the sender.
• L (Length) - length. Indicates the length of the frame so that the receiver can correctly predict the end of the frame.
• DSAP - Destination Service Access Point. 1 byte field. This is a service access point on the receiver's system that indicates where the memory buffers in the receiver's system should place the frame's data.
• SSAP - Source Service Access Point - also 1 byte field. This is an access point to the sender's system service that specifies where the sender's memory buffers should be placed on the sender's system.
• Control - Management. The field size is 1-2 bytes. This field indicates the type of service that is required for the data. Depending on what service you need to provide, the field can be either 1 or 2 bytes.
• data - data. The directly transmitted data. They can take a length of 46 to 1500 bytes.
• FCS - check for errors. Represents a checksum.

ethernet (read ethernet, from lat. aether - ether) - packet data transfer technology, mainly local
.

The Ethernet standards define wired connections and electrical signals at the physical layer, format
frames and media access control protocols - at the data link layer of the OSI model. ethernet basically
described by the IEEE 802.3 group standards. Ethernet has become the most common LAN technology in the middle
90s of the last century, replacing such obsolete technologies as Arcnet, FDDI and Token ring.

History of creation

Ethernet technology was developed along with many of the first projects of Xerox PARC Corporation.
It is generally accepted that Ethernet was invented on May 22, 1973, when Robert Metcalfe
wrote a memo to the head of PARC on the potential of Ethernet technology. But the legal right to
Metcalfe received the technology a few years later. In 1976 he and his assistant David Boggs
published a pamphlet titled "Ethernet: Distributed Packet-Switching For Local Computer Networks".

Metcalfe left Xerox in 1979 and founded 3Com to promote computers and local
computer networks (LAN). He managed to convince DEC, Intel and Xerox to work together and develop
Ethernet standard (DIX). This standard was first published on September 30, 1980. He started
rivalry with two large patented technologies, token ring and ARCNET, which were soon buried under the oncoming waves of Ethernet products. In the process of struggle, 3Com has become a major company in this industry.

The standard of the first versions (Ethernet v1.0 and Ethernet v2.0) states that as a transmission medium
coaxial cable is used, later it became possible to use twisted pair and optical
cable.

The reasons for switching to were:

• the ability to work in duplex mode;
• low cost of twisted pair cable;
• higher reliability of networks in the event of a cable failure;
• greater noise immunity when using a differential signal;
• the ability to power over the cable low-power nodes, such as IP phones (Power over Ethernet, POE standard);
• lack of galvanic connection (current flow) between network nodes. When using a coaxial cable in Russian conditions, where, as a rule, there is no grounding of computers, the use of a coaxial cable was often accompanied by a breakdown of network cards, and sometimes even a complete "burnout" of the system unit.

The reason for switching to optical cable was the need to increase the length of the segment without repeaters.

Access control method (for network on ) - Carrier Sense Multiple Access and
collision detection (CSMA/CD, Carrier Sense Multiple Access with Collision Detection), bit rate
data 10 Mbps, packet size from 72 to 1526 bytes, data encoding methods are described. Working mode
half-duplex, that is, the node cannot simultaneously transmit and receive information. Number of nodes in
one shared network segment is limited to a limit of 1024 workstations (specifications
physical layer can set more stringent restrictions, for example, to a thin coax segment
no more than 30 workstations can be connected, and no more than 100 workstations can connect to a thick coax segment). However
a network built on a single shared segment becomes inefficient long before reaching
node limit, mainly due to half-duplex operation.

In 1995, the IEEE 802.3u Fast Ethernet standard was adopted at 100 Mbps, and it became possible
full duplex operation. In 1997, the IEEE 802.3z Gigabit Ethernet standard was adopted at speeds
1000 Mbit/s for transmission over optical fiber and two years later for transmission over twisted pair.

Types of Ethernet

Depending on the data rate and transmission medium, there are several technology options.
Regardless of the method of transmission, the network protocol stack and programs work the same in almost all
all of the options listed below.

Most Ethernet cards and other devices support multiple baud rates,
using autodetection (autonegotiation) speed and duplex, to achieve the best
connections between two devices. If auto-detection does not work, the speed is adjusted to
partner, and the half-duplex transmission mode is turned on. For example, if the device has an Ethernet port
10/100 indicates that you can use 10BASE-T and 100BASE-TX technologies through it, and the port
Ethernet 10/100/1000 - Supports 10BASE-T, 100BASE-TX and 1000BASE-T standards.
Early modifications of Ethernet

• Xerox Ethernet - the original technology, speed 3Mbps, existed in two versions Version 1 and Version 2, the frame format of the latest version is still widely used.
• 10BROAD36 - not widely used. One of the first standards that allows you to work over long distances. Used a wideband modulation technology similar to that used
  in cable modems. A coaxial cable was used as the data transmission medium.
• 1BASE5, also known as StarLAN, was the first twisted-pair Ethernet technology. It worked at a speed of 1 Mbps, but did not find commercial use.

10 Mbps Ethernet

• 10BASE5, IEEE 802.3 (also called "Thick Ethernet") - the original development of technology with a data rate of 10 Mbps. Following an early IEEE standard, 50 ohm impedance (RG-8) coaxial cable is used, with a maximum segment length of 500 meters.
• 10BASE2, IEEE 802.3a (called "Thin Ethernet") - using RG-58 cable, with a maximum segment length of 185 meters, computers connected one to another, to connect the cable to the network
  The card needs a T-connector and the cable needs a BNC connector. Requires terminators on each
  end. For many years this standard was the main one for Ethernet technology.
• StarLAN 10 - The first development using twisted pair for data transmission at a speed of 10 Mbps.

Later evolved into the 10BASE-T standard.

Although it is theoretically possible to connect more than
two devices operating in simplex mode, this scheme is never used for Ethernet, in
different from working with . Therefore, all twisted-pair networks use a star topology,
while, networks on coaxial cable are built on a "bus" topology. Terminators for work on
twisted-pair cables are built into each device, and it is not necessary to use additional external terminators on the line.

• 10BASE-T, IEEE 802.3i - 4-wire twisted-pair cable (two twisted pairs) category-3 or category-5 is used for data transmission. The maximum segment length is 100 meters.
• FOIRL - (an acronym for the English Fiber-optic inter-repeater link). The base standard for Ethernet technology that uses an optical cable to transmit data. The maximum transmission distance without a repeater is 1 km.
• 10BASE-F, IEEE 802.3j - The main term for a family of 10 Mbps ethernet standards using optical cable at a distance of up to 2 kilometers: 10BASE-FL, 10BASE-FB and 10BASE-FP. Of these, only 10BASE-FL has become widespread.
• 10BASE-FL (Fiber Link) - Improved version of the FOIRL standard. The improvement concerned an increase in the length of the segment up to 2 km.
• 10BASE-FB (Fiber Backbone) - Now an unused standard, it was intended to combine repeaters into a backbone.
• 10BASE-FP (Fiber Passive) - A passive star topology that does not need repeaters - has never been used.

Fast Ethernet (Fast Ethernet, 100 Mbps)

• 100BASE-T is a generic term for standards that use . Segment length up to 100 meters. Includes 100BASE-TX, 100BASE-T4 and 100BASE-T2 standards.
• 100BASE-TX, IEEE 802.3u - development of the 10BASE-T standard for use in star topology networks. Category 5 twisted pair is used, in fact only two unshielded pairs of conductors are used, duplex data transmission is supported, distance up to 100 m.
• 100BASE-T4 is a category 3 twisted-pair standard. All four pairs of conductors are used, data transmission is in half duplex. Practically not used.
• 100BASE-T2 is a category 3 twisted-pair standard. Only two pairs of conductors are involved. Full duplex is supported, with signals propagating in opposite directions on each pair. The transfer rate in one direction is 50 Mbps. Practically not used.
• 100BASE-SX is a standard that uses multimode fiber. The maximum segment length is 400 meters in half duplex (for guaranteed collision detection) or 2 kilometers in full duplex.
• 100BASE-FX is a standard using single-mode fiber. The maximum length is only limited
  the value of attenuation in the optical cable and the power of transmitters, according to different materials from 2x to 10
  kilometers
• 100BASE-FX WDM is a standard using single mode fiber. The maximum length is only limited
  the amount of attenuation in the fiber optic cable and the power of the transmitters. There are two interfaces
  species, differ in the wavelength of the transmitter and are marked either with numbers (wavelength) or one Latin
  letter A(1310) or B(1550). Only paired interfaces can work in pairs: on the one hand, the transmitter
  at 1310 nm, and on the other - at 1550 nm.

• 1000BASE-T, IEEE 802.3ab is a standard using category 5e twisted pair. 4 pairs are involved in data transmission. Data transfer rate - 250 Mbit / s over one pair. The PAM5 coding method is used, the fundamental frequency is 62.5 MHz. Distance up to 100 meters
• 1000BASE-TX was created by the Telecommunications Industry Association (Telecommunications
  Industry Association, TIA) and published in March 2001 as "Physical Layer Specification
  duplex Ethernet 1000 Mbps (1000BASE-TX) Category 6 balanced cabling
  (ANSI/TIA/EIA-854-2001)" (Eng. "A Full Duplex Ethernet Specification for 1000 Mbis/s (1000BASE-TX)
  Operating Over Category 6 Balanced Twisted-Pair Cabling (ANSI/TIA/EIA-854-2001)"). standard, uses
  separate transmission and reception (one pair in each direction), which greatly simplifies the design
  transceiver devices. Another significant difference in 1000BASE-TX is the lack of circuitry
  digital interference compensation and return noise, resulting in complexity, power consumption
  and the price of processors becomes lower than those of 1000BASE-T processors. But, as a consequence, for
  stable operation of this technology requires a high quality cabling system, so 1000BASE-TX
  Can only use Category 6 cable. Based on this standard, almost no
  products, although 1000BASE-TX uses a simpler protocol than the 1000BASE-T standard and therefore may
  use simpler electronics.
• 1000BASE-X is a generic term for standards with pluggable GBIC or SFP transceivers.
• 1000BASE-SX, IEEE 802.3z is a standard using multimode fiber. Travel range
  signal without repeater up to 550 meters.
• 1000BASE-LX, IEEE 802.3z is a standard using single-mode fiber. Travel range
  signal without a repeater up to 5 kilometers.
• 
  
  used.
• 1000BASE-CX - standard for short distances (up to 25 meters) using twinax cable
  with a wave impedance of 75 ohms (each of the two waveguides). Superseded by 1000BASE-T and no longer
  used.
• 1000BASE-LH (Long Haul) is a standard using single-mode fiber. Travel range
  signal without a repeater up to 100 kilometers.

10 Gigabit Ethernet

The new 10 Gigabit Ethernet standard includes seven physical media standards for LAN, MAN and
wan. It is currently covered by the IEEE 802.3ae amendment and should be included in the next revision.
IEEE 802.3 standard.

• 10GBASE-CX4 - 10 Gigabit Ethernet technology for short distances (up to 15 meters), uses CX4 copper cable and InfiniBand connectors.
• 10GBASE-SR - 10 Gigabit Ethernet technology for short distances (up to 26 or 82 meters, depending on
  depending on cable type), multimode fiber is used. It also supports distances up to 300
  meters using a new multimode fiber (2000 MHz/km).
• 10GBASE-LX4 - uses wavelength multiplexing to support distances from 240 to 300 meters over multimode fiber. Also supports distances up to 10 kilometers when using single-mode
  fibers.
• 10GBASE-LR and 10GBASE-ER - these standards support distances up to 10 and 40 kilometers
  respectively.
• 10GBASE-SW, 10GBASE-LW and 10GBASE-EW - These standards use a physical interface compatible
  in terms of speed and data format with OC-192 / STM-64 SONET/SDH interface. They are similar to 10GBASE-SR standards,
  10GBASE-LR and 10GBASE-ER respectively as they use the same cable types and transmission distances.
• 10GBASE-T, IEEE 802.3an-2006 - adopted in June 2006 after 4 years of development. Uses
  shielded twisted pair. Distances - up to 100 meters.

• tutorial

• What is a collision domain?
• How many pairs are used for Ethernet and why?
• Which pairs are receiving and which are transmitting?
• What limits the length of a network segment?
• Why can't a frame be smaller than a certain value?

If you don’t know the answers to these questions, but you’re too lazy to read standards and serious literature on the topic, I ask under the cat.

Someone thinks that these are obvious things, others will say that it is a boring and unnecessary theory. Nevertheless, you can hear similar questions in interviews from time to time. My opinion: everyone who has to pick up an 8P8C “crimp” (this connector is usually mistakenly called RJ-45) needs to know about what will be discussed below. I do not pretend to academic depth, I will refrain from formulas and tables, and we will also leave linear coding behind. It will be mainly about copper wires, not about optics, because. they are more common in everyday life.

Ethernet technology describes two lower layers of the OSI model at once. Physical and channel. Further we will speak only about the physical, i.e. about how bits are transferred between two neighboring devices.

Ethernet technology is part of Xerox PARC's rich heritage. Early versions of Ethernet used coaxial cable as a transmission medium, but over time it was completely replaced by fiber optics and twisted pair. However, it is important to understand that the use of coaxial cable has largely determined how Ethernet works. The fact is that coaxial cable is a shared transmission medium. An important feature of a shared environment is that several interfaces can use it at the same time, but only one should transmit at a time. Using a coaxial cable, you can connect not only 2 computers to each other, but also more than two, without the use of active equipment. This topology is called tire. However, if at least two nodes on the same bus start simultaneously transmitting information, then their signals will overlap and the receivers of other nodes will not understand anything. Such a situation is called collision, and the part of the network in which the nodes compete for a common transmission medium - collision domain. In order to recognize a collision, the transmitting node constantly monitors the signals in the environment, and if its own transmitted signal differs from the observed one, a collision is recorded. In this case, all nodes stop transmitting and resume transmission after random time interval.

Collision domain diameter and minimum frame size

Now let's imagine what happens if, in the network shown in the figure, nodes A and C simultaneously start transmitting, but manage to finish it before they receive each other's signal. This is possible with a sufficiently short transmitted message and a sufficiently long cable, because, as we know from the school curriculum, the propagation speed of any signals is C=3*10 8 m/s at best. Because each of the transmitting nodes will receive an oncoming signal only after it has already finished transmitting its message - the fact that a collision has occurred will not be established by any of them, which means there will be no retransmission of frames. But node B at the input will receive the sum of the signals and will not be able to correctly receive any of them. In order to avoid such a situation, it is necessary to limit the size of the collision domain and minimum frame size. It is not difficult to guess that these quantities are directly proportional to each other. If the amount of transmitted information does not reach the minimum frame, then it is increased by a special pad field, the name of which can be translated as a placeholder.

Thus, the larger the potential size of the network segment, the more overhead is spent on the transmission of small data chunks. The developers of Ethernet technology had to find a middle ground between these two parameters, and the minimum frame size was set to 64 bytes.

Twisted pair and duplex operation

The question arises: where does the limit on the length of a segment for Ethernet over twisted pair come from if there is no shared medium? The thing is, the first twisted-pair networks used hubs. A hub (in other words, a multi-input repeater) is a device that has several Ethernet ports and broadcasts the received packet to all ports except the one from which this packet came. Thus, if the hub began to receive signals from two ports at once, then it did not know what to broadcast to the other ports, this was a collision. The same applied to the first Ethernet-networks using optics (10Base-FL).

Why then use a 4-pair cable, if out of 4 pairs only two are used? A reasonable question, and here are some reasons for doing this:

• A 4-pair cable is mechanically more reliable than a 2-pair cable.
• 4-pair cable does not have to be changed when moving to Gigabit Ethernet or 100BaseT4, which already use all 4 pairs
• If one pair is broken, you can use a free one instead and do not shift the cable
• Possibility to use Power over ethernet technology

Despite this, in practice they often use a 2-pair cable, connect 2 computers at once using one 4-pair cable, or use free pairs to connect a phone.

gigabit ethernet

Unlike its predecessors, Gigabit Ethernet always uses all 4 pairs for transmission at the same time. And in two directions at once. In addition, information is encoded not in two levels as usual (0 and 1), but in four (00,01,10,11). Those. the voltage level at any given moment encodes not one, but two bits at once. This is done in order to reduce the modulation frequency from 250 MHz to 125 MHz. In addition, a fifth level has been added to create code redundancy. It enables error correction at the reception. This type of coding is called five-level pulse-amplitude coding (PAM-5). In addition, in order to use all pairs simultaneously for receiving and transmitting, the network adapter subtracts its own transmitted signal from the total signal in order to receive the signal transmitted by the other side. Thus, full-duplex mode is realized on one channel.

Further more

10 Gigabit Ethernet is already in full use by providers, but is not used in the SOHO segment, tk. Apparently there is enough Gigabit Ethernet. 10GBE uses single and multimode fiber, with or without wavelength division multiplexing, copper cables with InfiniBand connectors, and twisted pair cables in the 10GBASE-T or IEEE 802.3an-2006 standard as the propagation medium.

40 Gigabit Ethernet (or 40GbE) and 100 Gigabit Ethernet (or 100GbE). The development of these standards was completed in July 2010. At the moment, leading network equipment manufacturers such as Cisco, Juniper Networks and Huawei are already developing and releasing the first routers supporting these technologies.

In conclusion, it is worth mentioning a promising technology Terabit Ethernet. Bob Metcalfe, the creator, suggested that the technology would be developed by 2015, and similarly said:

To implement 1TB Ethernet, many limitations must be overcome, including 1550nm lasers and 15GHz modulation. The future network needs new modulation schemes, as well as new fiber, new lasers, in general, everything is new.

UPD: Thanks to the Nickel3000 habrauser for suggesting that the connector, which I called RJ45 all my life, is actually 8P8C.
UPD2:: Thanks to user Wott for explaining why pins 1,2,3 and 6 are used.

Tags: Add tags

He and his assistant David Boggs published a pamphlet called Ethernet: Distributed Packet-Switching For Local Computer Networks.

Benefits of using twisted pair over coaxial cable:

The reason for switching to optical cable was the need to increase the length of the segment without repeaters.

Access control method (for a network on a coaxial cable) - multiple access with carrier sense and collision detection (CSMA / CD, Carrier Sense Multiple Access with Collision Detection), data transfer rate 10 Mbps, packet size from 72 to 1526 bytes, are described data encoding methods. The mode of operation is half-duplex, that is, the node cannot simultaneously transmit and receive information. The number of nodes in one shared network segment is limited to 1024 workstations (physical layer specifications may set more stringent restrictions, for example, no more than 30 workstations can connect to a thin coax segment, and no more than 100 to a thick coax segment). However, a network built on a single shared segment becomes inefficient long before reaching the limit of the number of nodes, mainly due to the half-duplex mode of operation.

MAC addresses

When designing the Ethernet standard, it was provided that each network card (as well as the built-in network interface) must have a unique six-byte number (MAC address) stitched into it during manufacture. This number is used to identify the sender and receiver of the frame, and it is assumed that when a new computer (or other device capable of networking) appears on the network, the network administrator will not have to configure the MAC address.

The uniqueness of MAC addresses is achieved by the fact that each manufacturer receives a range of sixteen million (2 ^ 24) addresses from the IEEE Registration Authority coordinating committee, and as the allocated addresses are exhausted, it can request a new range. Therefore, the manufacturer can be determined from the three most significant bytes of the MAC address. There are tables that allow you to determine the manufacturer by MAC address; in particular, they are included in programs like arpalert.

The MAC address is read once from the ROM when the network card is initialized, then all packets are generated by the operating system. All modern operating systems allow you to change it. For Windows, since at least Windows 98, it has changed in the registry. Some network card drivers made it possible to change it in the settings, but the change works absolutely for any cards.

Some time ago, when network card drivers did not allow you to change your MAC address, and alternative possibilities were not too well known, some Internet providers used it to identify a machine on a network when accounting for traffic. Microsoft Office programs since Office 97 have written the MAC address of the NIC to the document being edited as part of a unique GUID. . The MAC address of the router was transferred by the Mail.Ru agent to its server in clear text during login.

Types of Ethernet

Depending on the data transfer rate and the transmission medium, there are several technology options. Regardless of the method of transmission, the network protocol stack and programs work the same in almost all of the following options.

This section gives a brief description of all officially existing varieties. For some reason, in addition to the main standard, many manufacturers recommend using other proprietary media - for example, fiber optic cable is used to increase the distance between network points.

Most Ethernet cards and other devices support multiple data rates, using autonegotiation for speed and duplex to achieve the best possible connection between two devices. If auto-detection does not work, the speed is adjusted to the partner, and the half-duplex transmission mode is turned on. For example, the presence of a 10/100 Ethernet port in the device indicates that it can work through 10BASE-T and 100BASE-TX technologies, and the 10/100/1000 Ethernet port supports 10BASE-T, 100BASE-TX and 1000BASE- T.

Early modifications of Ethernet

• Xerox Ethernet- the original technology, speed 3 Mbps, existed in two versions Version 1 and Version 2, the frame format of the latest version is still widely used.
• 1BROAD36- not widely used. One of the first standards that allows you to work over long distances. Used wideband modulation technology, similar to that used in cable modems. A coaxial cable was used as the data transmission medium.
• 1BASE5- also known as StarLAN, was the first modification of Ethernet technology using twisted pair. It worked at a speed of 1 Mbps, but did not find commercial use.

10 Mbps Ethernet

• 10BASE5, IEEE 802.3 (also called "Thick Ethernet") - the original development of technology with a data transfer rate of 10 Mbps. Following an early IEEE standard, 50 ohm impedance (RG-8) coaxial cable is used, with a maximum segment length of 500 meters.
• 10BASE2, IEEE 802.3a (called "Thin Ethernet") - using RG-58 cable, with a maximum segment length of 185 meters, computers connected one to another, a T-connector is needed to connect the cable to the network card, and the cable must have a BNC connector . Requires terminators at each end. For many years this standard was the main one for Ethernet technology.
• StarLAN 10- The first development using twisted pair for data transmission at a speed of 10 Mbps. Later evolved into the 10BASE-T standard.

Although it is theoretically possible to connect more than two devices operating in simplex mode to one twisted pair cable (segment), this scheme is never used for Ethernet, in contrast to working with coaxial cable. Therefore, all twisted-pair networks use a star topology, while coaxial cable networks use a bus topology. Twisted-pair terminators are built into each device, and there is no need to use additional external terminators on the line.

• 10BASE-T, IEEE 802.3i - 4-wire twisted-pair cable (two twisted pairs) category-3 or category-5 is used for data transmission. The maximum segment length is 100 meters.
• FOIR- (acronym from English. Fiber-optic inter-repeater link ). The base standard for Ethernet technology that uses an optical cable to transmit data. The maximum transmission distance without a repeater is 1 km.
• 10BASE-F, IEEE 802.3j - The basic term for a family of 10 Mbps ethernet standards using optical cable at a distance of up to 2 kilometers: 10BASE-FL, 10BASE-FB and 10BASE-FP. Of these, only 10BASE-FL has become widespread.
• 10BASE-FL(Fiber Link) - An improved version of the FOIRL standard. The improvement concerned an increase in the length of the segment up to 2 km.
• 10BASE-FB(Fiber Backbone) - Now an unused standard, it was intended to combine repeaters into a backbone.
• 10BASE-FP(Fiber Passive) - A passive star topology that does not need repeaters - has never been used.

Fast Ethernet (Fast Ethernet , 100 Mbps)

• 100BASE-T- a general term for standards that use twisted pair as a data transmission medium. Segment length up to 100 meters. Includes 100BASE-TX, 100BASE-T4 and 100BASE-T2 standards.
• 100BASE-TX, IEEE 802.3u - development of the 10BASE-T standard for use in star topology networks. Category 5 twisted pair is used, in fact only two unshielded pairs of conductors are used, duplex data transmission is supported, distance up to 100 m.
• 100BASE-T4- a standard that uses category 3 twisted pair. All four pairs of conductors are used, data transmission is in half duplex. Practically not used.
• 100BASE-T2- standard using category 3 twisted pair. Only two pairs of conductors are used. Full duplex is supported, with signals propagating in opposite directions on each pair. The transfer rate in one direction is 50 Mbps. Practically not used.
• 100BASE-FX is a standard using single-mode fiber. The maximum segment length is 400 meters in half duplex (for guaranteed collision detection) or 2 kilometers in full duplex.
• 100BASE-SX is a standard using multimode fiber. The maximum length is limited only by the amount of attenuation in the optical cable and the power of the transmitters, according to different materials from 2 to 10 kilometers.
• 100BASE-FX WDM is a standard using single-mode fiber. The maximum length is limited only by the amount of attenuation in the fiber optic cable and the power of the transmitters. Interfaces are of two types, they differ in the transmitter wavelength and are marked either with numbers (wavelength) or with one Latin letter A (1310) or B (1550). Only paired interfaces can work in pairs: on the one hand, the transmitter is at 1310 nm, and on the other, at 1550 nm.

Gigabit Ethernet (Gigabit Ethernet, 1 Gbps)

10 Gigabit Ethernet (Ethernet 10G, 10 Gbps)

The new 10 Gigabit Ethernet standard includes seven physical media standards for LAN, MAN, and WAN. It is currently described by the IEEE 802.3ae amendment and should be included in the next revision of the IEEE 802.3 standard.

• 10GBASE-CX4- 10 Gigabit Ethernet technology for short distances (up to 15 meters), using CX4 copper cable and InfiniBand connectors.
• 10GBASE-SR- 10 Gigabit Ethernet technology for short distances (up to 26 or 82 meters, depending on cable type), using multimode fiber. It also supports distances up to 300 meters using the new multi-mode fiber (2000 MHz/km).
• 10GBASE-LX4- Uses wavelength multiplexing to support distances from 240 to 300 meters over multimode fiber. Also supports distances up to 10 kilometers when using single-mode fiber.
• 10GBASE-LR and 10GBASE-ER- these standards support distances up to 10 and 40 kilometers, respectively.
• 10GBASE-SW, 10GBASE-LW and 10GBASE-EW- These standards use a physical interface compatible in speed and data format with the OC-192/STM-64 SONET/SDH interface. They are similar to the 10GBASE-SR, 10GBASE-LR, and 10GBASE-ER standards, respectively, as they use the same cable types and transmission distances.
• 10GBASE-T, IEEE 802.3an-2006 - adopted in June 2006 after 4 years of development. Uses shielded twisted pair cable. Distances - up to 100 meters.
• 10GBASE-KR

Harting announced the creation of the world's first 10 Gigabit RJ-45 connector that does not require tools for installation - HARTING RJ Industrial 10G.

40 Gigabit and 100 Gigabit Ethernet

As observed by the 802.3ba Group, bandwidth requirements for computing and core applications are growing at different rates, which determines the need for two corresponding standards for the next generations of Ethernet - 40 Gigabit Ethernet (or 40GbE) and 100 Gigabit Ethernet (or 100GbE). Servers, high-performance computing clusters, blade systems, SANs and NAS are currently using 1GbE and 10GbE technologies, while in 2007 and 2008. there was a significant increase in the latter.

prospects

Terabit Ethernet (this is how Ethernet technology with a transfer rate of 1 TBit / s is simply called) became known in 2008 from a statement by the creator of Ethernet Bob Metcalfe at the OFC conference, who suggested that the technology would be developed by 2015, however, without expressing any or confidence, because for this you have to solve a lot of problems. However, in his opinion, the key technology that can serve the further growth of traffic will be one of those developed in the previous decade - DWDM.

“In order to implement 1TB Ethernet, many limitations must be overcome, including 1550nm lasers and 15GHz modulation. The future network needs new modulation schemes, as well as new fiber, new lasers, basically everything new,” Metcalfe said. - It is also unclear what network architecture will be required to support it. Perhaps the optical networks of the future will need to use vacuum core fiber or carbon fibers instead of quartz. Operators will need to implement more all-optical devices and free space (fiberless) optics. Bob Metcalf".

see also

Notes

Links

• IEEE 802.3 2008
• IEEE 802.3 2002