Editing Ethernet (section)

==Evolution==
{{IPstack}}
Ethernet has evolved to include higher bandwidth, improved [[medium access control]] methods, and different physical media. The [[multidrop]] coaxial cable was replaced with physical point-to-point links connected by [[Ethernet repeater]]s or [[Network switch|switches]].<ref>{{cite web |url=http://www.networkworld.com/article/2869883/lan-wan/evolution-of-ethernet.html |publisher=[[Network World]] |author=Jim Duffy |date=April 20, 2009 |access-date=January 1, 2016 |title=Evolution of Ethernet |archive-date=June 11, 2017 |archive-url=https://web.archive.org/web/20170611140149/http://www.networkworld.com/article/2869883/lan-wan/evolution-of-ethernet.html |url-status=dead }}</ref>

Ethernet stations communicate by sending each other [[data packet]]s: blocks of data individually sent and delivered. As with other IEEE 802 LANs, adapters come programmed with globally unique 48-bit [[MAC address]] so that each Ethernet station has a unique address.{{Efn|In some cases, the factory-assigned address can be overridden, either to avoid an address change when an adapter is replaced or to use [[locally administered address]]es.}} The MAC addresses are used to specify both the destination and the source of each data packet. Ethernet establishes link-level connections, which can be defined using both the destination and source addresses. On reception of a transmission, the receiver uses the destination address to determine whether the transmission is relevant to the station or should be ignored. A network interface normally does not accept packets addressed to other Ethernet stations.{{Efn|Unless it is put into [[promiscuous mode]].|name=promiscuous}}{{Efn|Of course bridges and switches will accept other addresses for forwarding the packet.}}

An EtherType field in each frame is used by the operating system on the receiving station to select the appropriate protocol module (e.g., an [[Internet Protocol]] version such as [[IPv4]]). [[Ethernet frame]]s are said to be ''self-identifying'', because of the EtherType field. Self-identifying frames make it possible to intermix multiple protocols on the same physical network and allow a single computer to use multiple protocols together.<ref>{{cite book |author=Douglas E. Comer |author-link=Douglas E. Comer |year=2000 |title=Internetworking with TCP/IP – Principles, Protocols and Architecture |edition=4th |publisher=Prentice Hall |isbn=0-13-018380-6}} 2.4.9 – Ethernet Hardware Addresses, p. 29, explains the filtering.</ref> Despite the evolution of Ethernet technology, all generations of Ethernet (excluding early experimental versions) use the same frame formats.<ref>{{cite web|author=Iljitsch van Beijnum|title=Speed matters: how Ethernet went from 3Mbps to 100Gbps... and beyond|url=https://arstechnica.com/gadgets/2011/07/ethernet-how-does-it-work/3/|website=[[Ars Technica]]|date=July 15, 2011|access-date=July 15, 2011|quote=All aspects of Ethernet were changed: its MAC procedure, the bit encoding, the wiring... only the packet format has remained the same.|archive-date=July 9, 2012|archive-url=https://web.archive.org/web/20120709015112/http://arstechnica.com/gadgets/2011/07/ethernet-how-does-it-work/3/|url-status=live}}</ref> Mixed-speed networks can be built using Ethernet switches and repeaters supporting the desired Ethernet variants.<ref>{{citation |url=http://www.lantronix.com/resources/networking-tutorials/fast-ethernet-tutorial/ |publisher=Lantronix |access-date=January 1, 2016 |title=Fast Ethernet Turtorial |date=December 9, 2014 |archive-date=November 28, 2015 |archive-url=https://web.archive.org/web/20151128172531/http://www.lantronix.com/resources/networking-tutorials/fast-ethernet-tutorial/ |url-status=live }}</ref>

Due to the ubiquity of Ethernet, and the ever-decreasing cost of the hardware needed to support it, by 2004 most manufacturers built Ethernet interfaces directly into [[PC motherboard]]s, eliminating the need for a separate network card.<ref>{{cite web |url=http://pcquest.ciol.com/content/search/showarticle.asp?artid=63428 |title=Motherboard Chipsets Roundup |publisher=PCQuest |date=November 1, 2004 |author=Geetaj Channana |quote=While comparing motherboards in the last issue we found that all motherboards support Ethernet connection on board. |access-date=October 22, 2010 |archive-url=https://web.archive.org/web/20110708154855/http://pcquest.ciol.com/content/search/showarticle.asp?artid=63428 |archive-date=July 8, 2011 |url-status=dead }}</ref>

=== Shared medium ===
[[File:10Base5transcievers.jpg|thumb|Older Ethernet equipment. Clockwise from top-left: An Ethernet transceiver with an in-line [[10BASE2]] adapter, a similar model transceiver with a [[10BASE5]] adapter, an [[Attachment Unit Interface|AUI]] cable, a different style of transceiver with 10BASE2 [[BNC connector|BNC]] T-connector, two 10BASE5 end fittings ([[N connector]]s), an orange ''[[vampire tap]]'' installation tool (which includes a specialized drill bit at one end and a socket wrench at the other), and an early model 10BASE5 transceiver (h4000) manufactured by DEC. The short length of yellow 10BASE5 cable has one end fitted with an N connector and the other end prepared to have an N connector shell installed; the half-black, half-grey rectangular object through which the cable passes is an installed vampire tap.]]

Ethernet was originally based on the idea of computers communicating over a shared coaxial cable acting as a broadcast transmission medium. The method used was similar to those used in radio systems,{{Efn|There are fundamental differences between wireless and wired shared-medium communication, such as the fact that it is much easier to detect collisions in a wired system than a wireless system.}} with the common cable providing the communication channel likened to the ''Luminiferous aether'' in 19th-century physics, and it was from this reference that the name ''Ethernet'' was derived.<ref name="Spurgeon 2000">{{cite book |title=Ethernet: The Definitive Guide |url=https://archive.org/details/ethernetdefiniti0000spur |url-access=registration |author=Charles E. Spurgeon |publisher=O'Reilly |isbn=978-1-56592-660-8 |year=2000}}</ref>

Original Ethernet's shared coaxial cable (the shared medium) traversed a building or campus to every attached machine. A scheme known as [[carrier-sense multiple access with collision detection]] (CSMA/CD) governed the way the computers shared the channel. This scheme was simpler than competing Token Ring or [[Token Bus]] technologies.{{Efn|In a CSMA/CD system packets must be large enough to guarantee that the leading edge of the propagating wave of a message gets to all parts of the medium and back again before the transmitter stops transmitting, guaranteeing that [[collisions]] (two or more packets initiated within a window of time that forced them to overlap) are discovered. As a result, the minimum packet size and the physical medium's total length are closely linked.}} Computers are connected to an [[Attachment Unit Interface]] (AUI) [[transceiver]], which is in turn connected to the cable (with [[thin Ethernet]] the transceiver is usually integrated into the network adapter). While a simple passive wire is highly reliable for small networks, it is not reliable for large extended networks, where damage to the wire in a single place, or a single bad connector, can make the whole Ethernet segment unusable.{{Efn|Multipoint systems are also prone to strange failure modes when an electrical discontinuity reflects the signal in such a manner that some nodes would work properly, while others work slowly because of excessive retries or not at all. See [[standing wave]] for an explanation. These could be much more difficult to diagnose than a complete failure of the segment.}}

Through the first half of the 1980s, Ethernet's [[10BASE5]] implementation used a coaxial cable {{convert|0.375|in}} in diameter, later called ''thick Ethernet'' or ''thicknet''. Its successor, [[10BASE2]], called ''thin Ethernet'' or ''thinnet'', used the [[RG-58]] coaxial cable. The emphasis was on making installation of the cable easier and less costly.<ref name=Hegering>{{cite book |author1=Heinz-Gerd Hegering |author2=Alfred Lapple |title=Ethernet: Building a Communications Infrastructure |publisher=Addison-Wesley |date=1993 |isbn=0-201-62405-2 |url-access=registration |url=https://archive.org/details/ethernetbuilding0000hege }}</ref>{{rp|57}}

Since all communication happens on the same wire, any information sent by one computer is received by all, even if that information is intended for just one destination.{{Efn|This ''one speaks, all listen'' property is a security weakness of shared-medium Ethernet, since a node on an Ethernet network can eavesdrop on all traffic on the wire if it so chooses.}} The network interface card interrupts the [[CPU]] only when applicable packets are received: the card ignores information not addressed to it.{{Efn|name=promiscuous}} Use of a single cable also means that the data bandwidth is shared, such that, for example, available data bandwidth to each device is halved when two stations are simultaneously active.<ref>{{citation |url=http://www.lantronix.com/resources/networking-tutorials/ethernet-tutorial-networking-basics/ |title=Ethernet Tutorial – Part I: Networking Basics |date=December 9, 2014 |publisher=Lantronix |access-date=January 1, 2016 |archive-date=February 13, 2016 |archive-url=https://web.archive.org/web/20160213014814/http://www.lantronix.com/resources/networking-tutorials/ethernet-tutorial-networking-basics/ |url-status=live }}</ref>

A collision happens when two stations attempt to transmit at the same time. They corrupt transmitted data and require stations to re-transmit. The lost data and re-transmission reduces throughput. In the worst case, where multiple active hosts connected with maximum allowed cable length attempt to transmit many short frames, excessive collisions can reduce throughput dramatically. However, a [[Xerox]] report in 1980 studied performance of an existing Ethernet installation under both normal and artificially generated heavy load. The report claimed that 98% throughput on the LAN was observed.<ref>{{cite journal| author1=Shoch, John F. |author2=Hupp, Jon A. | title = Measured performance of an Ethernet local network| journal=Communications of the ACM| volume = 23| issue = 12| pages = 711–721| publisher=ACM Press| date=December 1980| issn = 0001-0782
| doi = 10.1145/359038.359044|s2cid=1002624 | doi-access = free}}</ref> This is in contrast with [[token passing]] LANs (Token Ring, Token Bus), all of which suffer throughput degradation as each new node comes into the LAN, due to token waits. This report was controversial, as modeling showed that collision-based networks theoretically became unstable under loads as low as 37% of nominal capacity. Many early researchers failed to understand these results. Performance on real networks is significantly better.<ref>{{cite journal |author1=Boggs, D.R. |author2=Mogul, J.C. |author3=Kent, C.A. |name-list-style=amp |title=Measured capacity of an Ethernet: myths and reality |date=September 1988 |publisher=DEC WRL |url=http://www.hpl.hp.com/techreports/Compaq-DEC/WRL-88-4.pdf |journal= |access-date=December 20, 2012 |archive-date=March 2, 2012 |archive-url=https://web.archive.org/web/20120302125906/http://www.hpl.hp.com/techreports/Compaq-DEC/WRL-88-4.pdf |url-status=live }}</ref>

In a modern Ethernet, the stations do not all share one channel through a shared cable or a simple [[repeater hub]]; instead, each station communicates with a switch, which in turn forwards that traffic to the destination station. In this topology, collisions are only possible if station and switch attempt to communicate with each other at the same time, and collisions are limited to this link. Furthermore, the [[10BASE-T]] standard introduced a [[full duplex]] mode of operation which became common with [[Fast Ethernet]] and the de facto standard with [[Gigabit Ethernet]]. In full duplex, switch and station can send and receive simultaneously, and therefore modern Ethernets are completely collision-free.

<gallery class="center" caption="Comparison between original Ethernet and modern Ethernet" widths="250">
File:Bustopologie.png|The original Ethernet implementation: shared medium, collision-prone. All computers trying to communicate share the same cable, and so compete with each other.
File:HUB SWITCH 6.jpg|Modern Ethernet implementation: switched connection, collision-free. Each computer communicates only with its own switch, without competition for the cable with others.
</gallery>

===Repeaters and hubs===
[[Image:Network card.jpg|thumb|A 1990s [[Industry Standard Architecture|ISA]] [[network interface card]] supporting both coaxial-cable-based [[10BASE2]] ([[BNC connector]], left) and twisted-pair-based [[10BASE-T]] ([[8P8C]] connector, right)]]
{{Main|Ethernet hub}}

For signal degradation and timing reasons, coaxial [[Ethernet segment]]s have a restricted size.<ref>{{Cite web|url=https://kb.wisc.edu/ns/page.php?id=7829|title=Ethernet Media Standards and Distances|website=kb.wisc.edu|access-date=October 10, 2017|archive-date=June 19, 2010|archive-url=https://web.archive.org/web/20100619010200/https://kb.wisc.edu/ns/page.php?id=7829|url-status=live}}</ref> Somewhat larger networks can be built by using an [[Ethernet repeater]]. Early repeaters had only two ports, allowing, at most, a doubling of network size. Once repeaters with more than two ports became available, it was possible to wire the network in a [[star topology]]. Early experiments with star topologies (called ''Fibernet'') using [[optical fiber]] were published by 1978.<ref>{{cite journal |title= Fibemet: Multimode Optical Fibers for Local Computer Networks |author1= Eric G. Rawson |author2= Robert M. Metcalfe |journal= IEEE Transactions on Communications |date= July 1978 |volume= 26 |issue= 7 |pages= 983–990 |url= http://ethernethistory.typepad.com/papers/Fibernet.pdf |doi= 10.1109/TCOM.1978.1094189 |access-date= June 11, 2011 |archive-date= August 15, 2011 |archive-url= https://web.archive.org/web/20110815204821/http://ethernethistory.typepad.com/papers/Fibernet.pdf |url-status= live }}</ref>

Shared cable Ethernet is always hard to install in offices because its bus topology is in conflict with the star topology cable plans designed into buildings for telephony. Modifying Ethernet to conform to twisted-pair telephone wiring already installed in commercial buildings provided another opportunity to lower costs, expand the installed base, and leverage building design, and, thus, twisted-pair Ethernet was the next logical development in the mid-1980s.

Ethernet on unshielded twisted-pair cables (UTP) began with [[StarLAN]] at 1&nbsp;Mbit/s in the mid-1980s. In 1987 [[SynOptics]] introduced the first twisted-pair Ethernet at 10&nbsp;Mbit/s in a star-wired cabling topology with a central hub, later called [[LattisNet]].<ref name=VonBurg2003 /><ref name="Spurgeon 2000"/>{{rp|29}}<ref>{{cite book| title = The Triumph of Ethernet: technological communities and the battle for the LAN standard| author = Urs von Burg| publisher = Stanford University Press| year = 2001| url = https://books.google.com/books?id=ooBqdIXIqbwC&pg=PA175| isbn = 0-8047-4094-1| page = 175| access-date = September 23, 2016| archive-date = January 9, 2017| archive-url = https://web.archive.org/web/20170109135141/https://books.google.com/books?id=ooBqdIXIqbwC&pg=PA175| url-status = live}}</ref> These evolved into 10BASE-T, which was designed for point-to-point links only, and all termination was built into the device. This changed repeaters from a specialist device used at the center of large networks to a device that every twisted pair-based network with more than two machines had to use. The tree structure that resulted from this made Ethernet networks easier to maintain by preventing most faults with one peer or its associated cable from affecting other devices on the network.{{citation needed|date=April 2020|reason=OK, repeaters are required to deactivate ports that send excessive collisions, such as due to internal defects, or external wiring defects. That is an important part of this statement.}}

Despite the physical star topology and the presence of separate transmit and receive channels in the twisted pair and fiber media, repeater-based Ethernet networks still use half-duplex and CSMA/CD, with only minimal activity by the repeater, primarily generation of the [[jam signal]] in dealing with packet collisions. Every packet is sent to every other port on the repeater, so bandwidth and security problems are not addressed. The total throughput of the repeater is limited to that of a single link, and all links must operate at the same speed.<ref name="Spurgeon 2000"/>{{rp|278}}

=== Bridging and switching ===<!--[[Full-duplex Ethernet]] links here-->
[[File:Network switches.jpg|thumb|[[Patch cable]]s with [[patch field]]s of two Ethernet switches]]
{{Main|Network bridge|Network switch}}

While repeaters can isolate some aspects of [[Ethernet segment]]s, such as cable breakages, they still forward all traffic to all Ethernet devices. The entire network is one [[collision domain]], and all hosts have to be able to detect collisions anywhere on the network. This limits the number of repeaters between the farthest nodes and creates practical limits on how many machines can communicate on an Ethernet network. Segments joined by repeaters have to all operate at the same speed, making phased-in upgrades impossible.{{citation needed|date=April 2020}}

To alleviate these problems, bridging was created to communicate at the data link layer while isolating the physical layer. With bridging, only well-formed Ethernet packets are forwarded from one Ethernet segment to another; collisions and packet errors are isolated. At initial startup, Ethernet bridges work somewhat like Ethernet repeaters, passing all traffic between segments. By observing the source addresses of incoming frames, the bridge then builds an address table associating addresses to segments. Once an address is learned, the bridge forwards network traffic destined for that address only to the associated segment, improving overall performance. [[Broadcasting (networking)|Broadcast]] traffic is still forwarded to all network segments. Bridges also overcome the limits on total segments between two hosts and allow the mixing of speeds, both of which are critical to the incremental deployment of faster Ethernet variants.{{citation needed|date=April 2020}}

In 1989, [[Vanguard Managed Solutions|Motorola Codex]] introduced their 6310 EtherSpan, and [[Kalpana (company)|Kalpana]] introduced their EtherSwitch; these were examples of the first commercial Ethernet switches.{{Efn|The term ''switch'' was invented by device manufacturers and does not appear in the IEEE 802.3 standard.}} Early switches such as this used [[cut-through switching]] where only the header of the incoming packet is examined before it is either dropped or forwarded to another segment.<ref name="networkcomputing_2000">{{cite web |title=The 10 Most Important Products of the Decade |author=Robert J. Kohlhepp |date=October 2, 2000 |access-date=February 25, 2008 |publisher=Network Computing |url=http://www.networkcomputing.com/1119/1119f1products_5.html|archive-url=https://web.archive.org/web/20100105152318/http://www.networkcomputing.com/1119/1119f1products_5.html |archive-date=January 5, 2010}}</ref> This reduces the forwarding latency. One drawback of this method is that it does not readily allow a mixture of different link speeds. Another is that packets that have been corrupted are still propagated through the network. The eventual remedy for this was a return to the original [[store and forward]] approach of bridging, where the packet is read into a buffer on the switch in its entirety, its [[frame check sequence]] verified and only then the packet is forwarded.<ref name="networkcomputing_2000"/> In modern network equipment, this process is typically done using [[application-specific integrated circuit]]s allowing packets to be forwarded at [[wire speed]].{{citation needed|date=April 2020}}

When a twisted pair or fiber link segment is used and neither end is connected to a repeater, [[full-duplex]] Ethernet becomes possible over that segment. In full-duplex mode, both devices can transmit and receive to and from each other at the same time, and there is no collision domain.<ref>{{cite web |author=Nick Pidgeon |work=How Stuff Works |url=https://computer.howstuffworks.com/ethernet15.htm |title=Full-duplex Ethernet |date=April 2000 |access-date=February 3, 2020 |archive-date=June 4, 2020 |archive-url=https://web.archive.org/web/20200604085640/https://computer.howstuffworks.com/ethernet15.htm |url-status=live }}</ref> This doubles the aggregate bandwidth of the link and is sometimes advertised as double the link speed (for example, 200&nbsp;Mbit/s for Fast Ethernet).{{Efn|This is misleading, as performance will double only if traffic patterns are symmetrical.}} The elimination of the collision domain for these connections also means that all the link's bandwidth can be used by the two devices on that segment and that segment length is not limited by the constraints of collision detection.

Since packets are typically delivered only to the port they are intended for, traffic on a switched Ethernet is less public than on shared-medium Ethernet. <span id="switch_vulnerabilities">Despite this, switched Ethernet should still be regarded as an insecure network technology, because it is easy to subvert switched Ethernet systems by means such as [[ARP spoofing]] and [[MAC flooding]].</span>{{citation needed|date=April 2020}}<ref>{{Cite book|last1=Wang|first1=Shuangbao Paul|url=https://books.google.com/books?id=NFK_CyoyIGEC&pg=PT121|title=Computer Architecture and Security: Fundamentals of Designing Secure Computer Systems|last2=Ledley|first2=Robert S.|date=October 25, 2012|publisher=John Wiley & Sons|isbn=978-1-118-16883-7|language=en|access-date=October 2, 2020|archive-date=March 15, 2021|archive-url=https://web.archive.org/web/20210315204013/https://books.google.com/books?id=NFK_CyoyIGEC&pg=PT121|url-status=live}}</ref>

The bandwidth advantages, the improved isolation of devices from each other, the ability to easily mix different speeds of devices and the elimination of the chaining limits inherent in non-switched Ethernet have made switched Ethernet the dominant network technology.<ref>{{cite web |url=http://www.cisco.com/en/US/solutions/collateral/ns340/ns394/ns74/ns149/net_business_benefit09186a00800c92b9_ps6600_Products_White_Paper.html |quote=Respondents were first asked about their current and planned desktop LAN attachment standards. The results were clear—switched Fast Ethernet is the dominant choice for desktop connectivity to the network |title=Token Ring-to-Ethernet Migration |publisher=Cisco |access-date=October 22, 2010 |archive-date=July 8, 2011 |archive-url=https://web.archive.org/web/20110708160911/http://www.cisco.com/en/US/solutions/collateral/ns340/ns394/ns74/ns149/net_business_benefit09186a00800c92b9_ps6600_Products_White_Paper.html |url-status=live }}</ref>

===Advanced networking===
[[File:Coreswitch (2634205113).jpg|thumb|A core Ethernet switch]]

Simple switched Ethernet networks, while a great improvement over repeater-based Ethernet, suffer from single points of failure, attacks that trick switches or hosts into sending data to a machine even if it is not intended for it, scalability and security issues with regard to [[switching loop]]s, [[broadcast radiation]], and [[multicast]] traffic.{{citation needed|date=April 2020}}

Advanced networking features in switches use [[Shortest Path Bridging]] (SPB) or the [[Spanning Tree Protocol]] (STP) to maintain a loop-free, meshed network, allowing physical loops for redundancy (STP) or load-balancing (SPB). Shortest Path Bridging includes the use of the [[link-state routing protocol]] [[IS-IS]] to allow larger networks with shortest path routes between devices.

Advanced networking features also ensure port security, provide protection features such as MAC lockdown<ref>{{cite web |url=https://www.techrepublic.com/blog/it-security/lock-down-cisco-switch-port-security-88196/ |title=Lock down Cisco switch port security |author=David Davis |date=October 11, 2007 |access-date=April 19, 2020 |archive-date=July 31, 2020 |archive-url=https://web.archive.org/web/20200731010910/https://www.techrepublic.com/blog/it-security/lock-down-cisco-switch-port-security-88196/ |url-status=live }}</ref> and broadcast radiation filtering, use [[VLAN]]s to keep different classes of users separate while using the same physical infrastructure, employ [[multilayer switch]]ing to route between different classes, and use [[link aggregation]] to add bandwidth to overloaded links and to provide some redundancy.{{citation needed|date=April 2020}}

In 2016, Ethernet replaced [[InfiniBand]] as the most popular system interconnect of [[TOP500]] supercomputers.<ref>{{cite web |url=https://www.top500.org/lists/top500/2016/06/highlights/ |title=HIGHLIGHTS – JUNE 2016 |quote=InfiniBand technology is now found on 205 systems, down from 235 systems, and is now the second most-used internal system interconnect technology. Gigabit Ethernet has risen to 218 systems up from 182 systems, in large part thanks to 176 systems now using 10G interfaces. |date=June 2016 |access-date=February 19, 2021 |archive-date=January 30, 2021 |archive-url=https://web.archive.org/web/20210130100950/https://top500.org/lists/top500/2016/06/highlights/ |url-status=live }}</ref>