Which company developed the first ethernet network standard. What is Ethernet - technology features and operating principles

Ethernet is the most common technology for organizing local networks. Ethernet standards describe the implementation of the first two layers of the OSI model - wired connections and electrical signals (physical layer), as well as data block formats and network access control protocols (link layer). Let's start with the idea behind Ethernet. The name Ethernet comes from two English words - ether (ether) and net (network). Ethernet uses the concept of a shared airwave. Each PC sends data into this ether and indicates to whom it is addressed. Data can reach all PCs on the network, but only the PC for which it is intended processes it. Other PCs ignore other people's data. This work is similar to broadcasting on radio stations. All radio stations broadcast their broadcasts into a common electromagnetic field - radio air. Your radio receives electromagnetic signals from all stations. But you don’t listen to everything at once, but to the station you need.

– the best price/speed/capability ratio compared to all other technologies

– the ability to use internal network resources at high speed without restrictions or additional payment

– no initial investment for equipment in the apartment

– independence from either the availability of a landline telephone/television cable or their technical condition

– there are many additional devices for easy connection on one line of several computers and/or other devices (game consoles, IP surveillance cameras, etc.), since this is the main and most common network technology today and in the foreseeable future.

– the opportunity to get high-speed tariffs sufficient for the most advanced users

– the ability to quickly, without problems and additional costs for equipment, change providers

– the need to introduce additional cable into the apartment

Ethernet is not one, but a whole family of standards with different user characteristics.

If we take the data transfer speed and the maximum possible distance between two nodes (network diameter) as the basis for comparing these standards, we get the following comparison table:

Table 1.6.1 Types Ethernet technologies

Ethernet standard type	Data transfer rate	Maximum network diameter

	100 Mbit/s
Gigabit Ethernet	1000 Mbit/s

First, let's look at the principle of building local networks based on the historically first version of Ethernet (10 Mbit/s), which appeared in the late 70s as a standard of three companies - Digital, Intel, Xerox.

This technology, like Fast Ethernet and Gigabit Ethernet technologies, is based on the concept of a shared medium: each node receives everything that is transmitted over the network; Only one node performs the transmission; the rest wait for a pause to begin their own transmission.

10G Ethernet technology is based on a different principle: information is not “scattered” throughout the network, but is purposefully “pushed” from node to node towards its destination. Routers are responsible for moving data through such a network. They determine the neighboring node to which the information packet needs to be moved to bring it closer to its destination. Such networks are called packet-switched networks.

Figure 1.6.1 shows the diagram Ethernet networks on a coaxial cable. The cable segment at the ends is equipped with terminators (“plugs”) to absorb the propagated signal (in the figure, the terminators are drawn as black squares).

Figure 1.6.1 Ethernet network diagram

The cable connects the network adapters of computers using a T-shaped connector.

How Ethernet works.

Any participant can send a message to the network, but only when it is “quiet” - there is no other transmission.

For example, node 2 (see figure above) listens to the network, and starts transmitting, starting with the sender and recipient addresses (“computer 2 transmits a message for computer 4”).

The transmission travels along the cable in both directions (absorbed by terminators at the ends), and all participants hear it (including the sender himself).

Everyone except computer 4 ignores the transmitted data, having detected someone else's recipient address, and computer 4 receives the data completely.

It is clear that with this transmission method it is impossible to allow a long-term takeover of the network by one node. If computer 2 decides to send a large file to computer 4, all other network participants will not soon have the opportunity to start the transfer.

For this reason, messages are transmitted divided into packets (in Ethernet technology they are called frames). The packet length ranges from 64 to 1518 bytes.

Having transmitted one packet, the node interrupts work for a while, and if the network is “quiet”, it sends the next packet. But another node can take advantage of the pause and begin its transmission session. Thus, all nodes share one medium (cable), having equal opportunities to send information packets to the network.

I recently visited an internet forum where people were discussing their 1Gbps fiber internet connections. "Lucky them!" - I thought. But is it really about luck? If you notice that instead of 1 Gbps you are getting about 80 Mbps, or even less, the problem may be the wrong Ethernet cable.

In this article we will tell you how to choose the right Ethernet cable for maximum Internet connection speed.

WiFi vs Ethernet

Let's immediately find out that an Ethernet cable provides more high speeds Internet connections than Wi-Fi. Yes, demon wired network– this is very convenient, but if you want to get maximum speed Internet connection, then you should use an Ethernet cable.

Ethernet to the rescue!

Naturally, if you have a wired network and very fast broadband Internet, you don't want to use a 100 Mbps (Fast Ethernet) connection between your computer and your ISP modem. That would be stupid! You need gigabit internet.

All you need is to connect all your home devices with inexpensive Ethernet cables Cat 6, and use cheap gigabit switches as "nodes" to connect your devices.

My home network looks like this:

Pretty simple, isn't it?

The orange line is a Cat 6 Ethernet cable. You simply connect computers, routers, laptops using these cables and everything “just works”.

However, you should note that some laptops come with cheap built-in Fast Ethernet adapters that offer connection speeds of no more than 100 Mbps. If you have this situation with your computer, buy a gigabit USB-ethernet adapter.

But which switches and Ethernet cables should you buy?

This is also a pretty easy question.

For Ethernet switches, you need a quality “Gigabit Ethernet Switch”. We recommend purchasing the 8-port D-Link Gigabit DGS-108, which is perfect for home use.

This switch is very easy to use: when you plug in the Ethernet cable and the connector flashes green, then it works at 1 gigabit speed. If the indicator is orange, the speed is only 10 or 100 Mbit/s. This way you can determine which Ethernet adapter used in your computer, as we already discussed above.

When it comes to Ethernet cables, you just need to make sure that you are using Cat 6 (Category 6). Ethernet cables usually have a category printed on them, such as:

Please note that there are other types of Ethernet cables, such as Cat 5, Cat 5e, Cat 6a, etc. Any cable that says Cat 6 is a great option for our situation (regardless of the letter at the end, if any). You should not buy Cat 5 Ethernet cables because they are designed to operate on networks less than 1 Gbps.

By the way, the connectors on Ethernet cables do not play a special role in the quality and speed of the signal. Four twisted pairs of wires inside the cable play much higher value. The higher the category, the faster the cable will transmit data. This is why you should use Cat 6 or higher. Cat 6 is for Gigabit Ethernet!

Also, you don't have to worry about shielding if you buy a ready-made cable. Just make sure it's Cat 6 and you're good to go!

We've prepared some tips and notes on using Ethernet cables throughout your home:

Do not unwind the network cable;
Do not pinch the cable in the doors;
Do not bend the cable at a right angle; round it at the corners.

Cat 6 Ethernet cable is a little stronger than others because it has a plastic core that accommodates twisted pairs of wires. But you still shouldn't abuse the strength of the cable. The more you squeeze the cable, the more the wires inside will move, and the lower the data transfer speed will be.

Using several simple tips, you can make your own home network as quickly as possible. A 1 Gbps Internet connection is not a problem, of course, if your Internet provider offers such fast broadband.

Ethernet—most networks these days are based on it. There are a large number of technologies that allow you to connect computers into a network. Each of them was developed in different time and is designed to solve a specific problem.

Ethernet technology covers two lower levels OSI models. Physical and channel levels. Further we will talk only about the physical layer of the OSI model, i.e. about how data bits are transferred between two neighboring devices.

Currently, technology is used to build local networks FastEthernet, which is a new implementation of the technology Ethernet.

What is Ethernet

This technology was developed in 1970 by the Palo Alto Research Center, which belongs to the Xerox Corporation, and in 1980, the IEEE 802.3 specification was adopted on its basis.

The basic operating principle used in this technology is as follows. To start transmitting data on the network, network adapter The computer “listens” the network for any signal. If it is not present, then the adapter begins data transmission, but if there is a signal, then the transmission is delayed for a certain time interval. Time exclusive use shared medium by one node is limited by the transmission time of one frame.

Frame - it is a unit of data exchanged between computers on an Ethernet network. The frame has a fixed format and, along with the data field, contains various service information, such as the recipient's address and the sender's address. After the sender adapter has placed the frame on the network, all network adapters begin to receive it. Each adapter analyzes the frame, and if the address matches their own device address (MAC address), the frame is placed in the internal buffer of the network adapter, but if it does not match, it is ignored.

In the event that two or more adapters, having “listened” to the network, begin to transmit data, a collision (collision). The adapters, having detected a collision, stop data transmission, and then, after “listening” to the network again, repeat data transmission at different intervals.

? NOTE. To receive a data packet that is destined for a particular adapter, it must accept all packets that appear on the network.

This method of accessing the data transmission medium is called CSMA/ CD(carrier-sense multiple access/collision detection) — multiple access with carrier detection.

What is Ethernet - collisions

As follows from the above, when large number computers on the network. and with intensive information exchange, the number of collisions grows very quickly. and as a result, network throughput decreases. It is possible that the throughput may drop to zero. But even in a network where the average load does not exceed the recommended one. This is 30-40% of the total bandwidth, the transmission speed is 70-80% of the nominal.

However, currently this problem almost decided. Because they have developed devices capable of dividing data streams between those computers for which this data is intended. In other words, traffic between ports connected to the sending and receiving network adapters is isolated from other ports and adapters. Such devices are called switches (switch).

There are various implementations of this technology - Ethernet, Fast Ethernet, Gigabit Ethernet. For example, they can provide data transfer rates of 10, 100 and 1000 Mbit/s, respectively.

The IEEE 802.3 standard contains several specifications that differ in topology and type of cable used. For example, 10 BASE-5 uses thick coaxial cable. 10 BASE-2 is a thin cable. And 10 BASE-F, 10 BASE-FB, 10 BASE-FL and FOIRL use optical cable. The most popular specification is IEEE 802.3 100BASE-TX. In which a cable based on unshielded twisted pairs with RJ-45 connectors is used to organize the network.

Ethernet Network Implementations

The Ethernet specifications listed above can be described as follows. The first number in the specification name indicates the maximum data transfer rate. For example, “10” indicates a signal transmission rate of 10 Mbit/s. “Base” means the use of Baseband technology in the standard. B aseband- This is narrowband transmission. With this method of transmitting data over a cable, every bit of data is encoded. It is encoded by a separate electrical or light pulse. In this case, the entire cable is used as one communication channel. Those. simultaneous transmission of two signals is not possible.

Originally, the last section in the specification title was intended to display maximum length. Cable segment lengths in hundreds of meters. This is without hubs or switches. However, for convenience and a more complete definition of the essence of the standard, everything has been simplified. And now the numbers in its name have been replaced with the letters T and F. Where T stands for twistedpair— twisted pair, and F denotes optical fiber.

Thus, nowadays one can find networks based on the following specifications:

10Base-2 - 10 MHz Ethernet on coaxial cable with a resistance of 50 Ohms, baseband. 10Base-2 is known as "thin Ethernet";
10Base-5 - 10MHzEthernet on a standard (thick) coaxial cable with a resistance of 50 Ohms, baseband;
10Base-T - 10MHz Ethernet over twisted pair cable;
100 Base-TX - 100MHz Ethernet over twisted pair cable.

Very significant advantage The different Ethernet options are mutually compatible. One that allows you to use them together on the same network. And in some cases, without even changing the existing cable system.

FULL DUPLEX MODE

The Fast Ethernet technology standard also includes recommendations. Recommendations for enabling full duplex operation (full— duplexmode) when connecting the network adapter to the switch. Or when directly connecting switches to each other.

The essence of full-duplex mode is the ability to simultaneously transmit and receive data over two channels. Tx (channel from transmitter to receiver) and Rx (channel from receiver to transmitter). And at the same time, the transmission speed doubles and reaches 200 Mbit/s.

On this moment almost all manufacturers network equipment state the following. That their devices provide full-duplex operation. However, due to different interpretations of the standard, in particular how to manage the flow of personnel. It is not always possible to achieve correct operation of these devices. And also good speed performance.

access with carrier sense and collision detection (Carrier Sense Multiply Access with Collision Detection - CSMA/CD). All computers on the network have access to a common bus through the network adapter built into each computer using semi duplex mode transfers. The diagram for connecting computers via coaxial cable is shown in Fig. 6.1.

Rice. 6.1.

Stations on a traditional local network Ethernet can be connected together using a physical bus or a star topology, but a logical topology is always a bus topology. By this we mean that the medium (channel) is shared among stations and only one station at a time can use it. It is also assumed that all stations receive the frame sent by the station (broadcast). The addressed destination saves the frame while the others discard it. In this situation, how can we ensure that two stations are not using the medium at the same time? Answer: if their frames collide with each other. CSMA/CD is designed to solve this problem according to the following principles:

Each station has an equal right to the medium (collective access).
Each station having a frame to send first "listens" (monitors) the medium. If there is no data in the medium, the station can start transmitting (carrier frequency tracking).
It may happen that two stations monitoring the medium find that it is not busy and start sending data. In this case, a conflict arises, called a collision.

The protocol forces the station to continue monitoring the line after transmission has begun. If there is a collision, then all stations detect it, each transmitting station sends a failure signal to destroy the line data, and then waits a different random time each time to try again. Random times prevent data from being resent at the same time. Before starting transmission, the node must ensure that the carrier medium is not busy, which is indicated by the absence of carrier frequency. If the medium is free, then the node has the right to start transmitting a frame of a certain format. Suppose that node 2 needs to transmit a frame to node N. Having discovered that the medium is free, node 2 begins transmitting the frame (Fig. 6.2), which is preceded by preamble, consisting of 7 bytes of the form 10101010, and a byte Start of Frame Delimiter - SFD type 10101011. These combinations are needed by the receiver to enter bit and frame synchronization with the transmitter. The frame ends with a frame control sequence field (FCS - Frame Check Sequence) 4 bytes long (not shown in Fig. 6.2). The transmitter signals propagate in both directions along the cable, and all nodes recognize the start of frame transmission. Only node N recognizes its own address(destination MAC address) at the beginning of the frame and writes its contents to its buffer for processing. From the received frame, the source address (source MAC address) is determined to which the response frame should be sent. The recipient of the packet at layer 3 is determined according to the field Protocol Type: value 0x0800 - IP module address, 0806 - ARP module address. Minimum and maximum field length values for protocols upper levels- 46 and 1500 bytes, respectively. The order of frame bit transmission is: left to right / bottom to top (Fig. 6.2), numbers indicate the lengths of the frame fields in bytes.

Any node, if there is a frame to transmit and a busy medium, is forced to wait for its release. A sign of the end of transmission is the disappearance of the carrier frequency. After the end of frame transmission, all nodes must withstand a technological pause of 9.6 μs to bring the network adapters into the initial state and prevent the same node from taking over the environment again.

Rice. 6.2.

Sometimes situations arise when one node has already started transmitting, but the other node has not yet had time to detect this and also begins transmitting its frame. This situation of free medium being captured by more than one node is called collision. The collision resolution mechanism is as follows (Fig. 6.3):

Rice. 6.3.

If the level of the received signal does not exceed the threshold value, then the node continues transmission, but if it exceeds it, then the node stops transmitting the frame and sends a special 32-bit jam combination (collision signal) to the network with an ad hoc sequence, which simply leads to an increase in the signal level in the local network due to an increase in pulse amplitude Manchester area code total signal. After this, the node that detected the collision pauses randomly and can then try to transmit the frame again. The number of retries cannot exceed 16. If, after the 16th attempt, the frame causes a collision, it is discarded. At large quantities nodes, the probability of collision increases, and throughput Ethernet network drops because The network is increasingly busy processing collisions and dropping frames. Three factors determine how CSMA/CD works: the minimum frame length, transmission speed data and conflict domain.

You have to wait for the station certain time to ensure there is no data on the line - this time is equal to the minimum frame length divided by transmission speed(the time it takes to transmit a frame of minimum length), and is proportional to the time it takes for the first bit to travel the maximum network distance (collision domain). In other words, we have:

Minimum Frame Length/Bit Rate is proportional to Collision Domain/Propagation Rate

In traditional Local network Ethernet, minimum frame length is 520 bits, transmission speed- 10 Mbit/s, propagation speed is almost equal to the speed of light, and the conflict domain is about 2500 meters.

System administration

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 size?

If you don’t know the answers to these questions, and are too lazy to read standards and serious literature on the topic, please refer to the cat.

Some people think that these are obvious things, others will say that it is a boring and unnecessary theory. Nevertheless, during interviews you can periodically hear such questions. My opinion: everyone who has to pick up an 8P8C “crimp” (this connector is usually mistakenly called RJ-45) needs to know what will be discussed below. I don’t pretend to have any academic depth, I’ll refrain from formulas and tables, and we’ll also leave linear coding behind. We will talk mainly about copper wires, not about optics, because... they are more widespread in everyday life.

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

Ethernet technology is part of the rich heritage of the Xerox PARC Research Center. Early versions Ethernet used coaxial cable as a transmission medium, but over time it was completely replaced by optical fiber and twisted pair cable. However, it is important to understand that the use coaxial cable largely determined the principles of Ethernet operation. The fact is that coaxial cable is a shared transmission medium. Important Feature shared medium: several interfaces can use it simultaneously, 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 using active equipment. This topology is called tire. However, if at least two nodes on the same bus start transmitting information simultaneously, their signals will overlap each other and the receivers of other nodes will not understand anything. This situation is called collision, and part of the network in which nodes compete for general environment transfers - 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 detected. In this case, all nodes stop transmitting and resume transmission through random time interval.

Collision domain diameter and minimum frame size

Now let's imagine what will happen if, in the network shown in the figure, nodes A and C start transmitting at the same time, but manage to finish it before they receive each other's signal. This is possible with a short enough transmitted message and a sufficiently long cable, because as we know from school curriculum, the speed of propagation of any signals in best case scenario is C=3*10 8 m/s. Because each of the transmitting nodes will receive a counter 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 will receive a sum of signals at the input and will not be able to correctly receive any of them. In order to prevent this situation from happening, 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. In case the volume transmitted information does not reach the minimum frame, it is increased due to the 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 transferring data chunks small size. Ethernet technology developers had to look for a middle ground between these two parameters, and minimum size frame size was set to 64 bytes.

Twisted pair and full duplex operation

Twisted pair as a transmission medium differs from coaxial cable in that it can only connect two nodes and uses separate media to transmit information in different directions. One pair is used for transmitting (1.2 pins, typically orange and white-orange wires) and one pair for receiving (3.6 pins, typically green and white-green wires). On active network equipment it’s the other way around. It is not difficult to notice that the central pair of contacts is missing: 4, 5. This pair was deliberately left free; if you insert an RJ11 into the same socket, it will occupy exactly the free contacts. This way you can use one cable and one socket for LAN and, for example, telephone. The pairs in the cable are selected in such a way as to minimize the mutual influence of signals on each other and improve the quality of communication. The wires of one pair are twisted together so that the influence of external noise on both wires in the pair is approximately the same.
To connect two devices of the same type, for example two computers, a so-called crossover cable is used, in which one pair connects contacts 1,2 of one side and 3,6 of the other, and the second vice versa: 3,6 contacts of one side and 1 ,2 other. This is necessary in order to connect the receiver to the transmitter; if you use a straight cable, you get a receiver-receiver, transmitter-transmitter. Although now this only matters if you work with some archaic equipment, because... Almost all modern equipment supports Auto-MDIX - a technology that allows the interface to automatically determine which pair is receiving and which is transmitting.

The question arises: where does the limitation on the segment length of Ethernet over twisted pair come from if there is no shared medium? The thing is, the first networks built on twisted pair cables 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 the 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 remaining ports; it was a collision. The same applied to the first Ethernet networks using optics (10Base-FL).

Why then use a 4-pair cable if only two of the 4 pairs are used? A reasonable question, and here are some reasons to do so:

A 4-pair cable is mechanically more reliable than a 2-pair cable.
The 4-pair cable will not have to be changed when switching to Gigabit Ethernet or 100BaseT4, which already uses all 4 pairs
If one pair is broken, you can use a free one instead and not re-wire 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 telephone.

Gigabit Ethernet

Unlike its predecessors, Gigabit Ethernet always uses all 4 pairs for transmission simultaneously. 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 was 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 makes it possible to correct errors during reception. This type of coding is called five-level pulse amplitude coding (PAM-5). Moreover, in order to use all pairs simultaneously To receive and transmit, the network adapter subtracts its own transmitted signal from the total signal to obtain the signal transmitted by the other side. In this way, full-duplex mode is implemented over one channel.

Further more

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

40 Gigabit Ethernet (or 40GbE) and 100 Gigabit Ethernet (or 100GbE). The development of these standards was completed in July 2010. IN currently 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 promising technology Terabit Ethernet. Bob Metcalfe, the creator, suggested that the technology would be developed by 2015 and also said:

To realize 1 TB/s Ethernet, many limitations must be overcome, including 1550 nm lasers and 15 GHz modulation. For future network we need new modulation schemes, as well as new optical fiber, new lasers, in general, everything new

UPD: Thanks to the hubbrowser for telling me that the connector that I have called RJ45 all my life is actually 8P8C.
UPD2:: Thanks to the user for explaining why pins 1,2,3 and 6 are used.