Professional Documents
Culture Documents
1: O Que Ethernet ?
Ethernet
Origem: Wikipdia, a enciclopdia livre.
Protocolo
1 Histria
2 Descrio geral
4 Hubs Ethernet
7 Variedades de Ethernet
9 10 Mbit/s Ethernet
10 Fast Ethernet
11 Gigabit Ethernet
13 Padres relacionados
14 Ver tambm
15 Referncias
16 Ligaes externas
Uma placa de rede Ethernet tpica com conectores BNC(esquerda) e RJ-45 (centro).
Ethernet baseada na ideia de pontos da rede enviando mensagens, no que essencialmente semelhante
a um sistema de rdio, cativo entre um cabo comum ou canal, s vezes chamado de ter (no original, ether).
Isto uma referncia oblqua ao ter luminfero, meio atravs do qual os fsicos do sculo XIX acreditavam
que a luz viajasse.1
Cada ponto tem uma chave de 48 bits globalmente nica, conhecida como endereo MAC, para assegurar
que todos os sistemas em uma ethernet tenham endereos distintos.
Tem sido observado que o trfego Ethernet tem propriedades de auto-similaridade, com importantes
consequncias para engenharia de trfego de telecomunicaes.
Os padres atuais do protocolo Ethernet so os seguintes:
Como todas as comunicaes aconteciam em um mesmo fio, qualquer informao enviada por um
computador era recebida por todos os outros, mesmo que a informao fosse destinada para um destinatrio
especfico. A placa de interface de rede descarta a informao no endereada a ela, interrompendo a CPU
somente quando pacotes aplicveis eram recebidos, a menos que a placa fosse colocada em seu modo
de comunicao promscua. Essa forma de um fala e todos escutam definia um meio de compartilhamento
de Ethernet de fraca segurana, pois um nodo na rede Ethernet podia escutar s escondidas todo o trfego
do cabo se assim desejasse. Usar um cabo nico tambm significava que a largura de banda (bandwidth)
era compartilhada, de forma que o trfego de rede podia tornar-se lentssimo quando, por exemplo, a rede e
os ns tinham de ser reinicializados aps uma interrupo eltrica.
O quadro Ethernet verso 2 ou quadro Ethernet II, chamado quadro DIX (iniciais de DEC, Intel,
e Xerox). o mais comum atualmente, j que muitas vezes usado diretamente pelo Protocolo Internet.
Os tipos diferentes de quadro tm formatos e valores de MTU diferentes, mas podem coexistir no mesmo
meio fsico.
A Ethernet Verso 1 original da Xerox tinha um campo de comprimento de 16 bits, embora o tamanho
mximo de um pacote fosse 1500 bytes. Esse campo de comprimento foi logo reusado na Ethernet Verso 2
da Xerox como um campo de rtulo, com a conveno de que valores entre 0 e 1500 indicavam o uso do
formato Ethernet original, mas valores maiores indicavam o que se tornou conhecido como um EtherType, e
o uso do novo formato de quadro. Isso agora suportado nos protocolos IEEE 802 usando o header SNAP.
O IEEE 802.x definiu o campo de 16 bits aps o endereo MAC como um campo de comprimento de novo.
Como o formato de quadros do Ethernet I no mais usado, isso permite ao software determinar se um
quadro do Ethernet II ou do IEEE 802.x, permitindo a coexistncia dos dois padres no mesmo meio fsico.
Todos os quadros 802.x tm um campo LLC. Examinando o campo LLC, possvel determinar se ele
seguido por um campo SNAP.
As variantes 802.x de Ethernet no so de uso geral em redes comuns. O tipo mais comum usado hoje a
Ethernet Verso 2, j que usada pela maioria das redes baseadas noProtocolo da Internet, com
seu EtherType setado em 0x0800. Existem tcnicas para encapsular trfego IP em quadros IEEE 802.3, por
exemplo, mas isso no comum.
Xerox Ethernet -- a implementao original de Ethernet, que tinha 2 verses, Verso 1 e Verso 2,
durante seu desenvolvimento. O formato de frame da verso 2 ainda est em uso comum.
10BASE5 (tambm chamado Thicknet) -- esse padro antigo da IEEE usa um cabo coaxial simples
em que voc conseguia uma conexo literalmente furando o cabo para se conectar ao ncleo. um
sistema obsoleto, embora devido a sua implantao amplamente difundida antigamente, talvez ainda
possa ser utilizado por alguns sistemas.
10BROAD36 -- Obsoleto. Um padro antigo permitindo a Ethernet para distncias mais longas.
Utilizava tcnicas de modulao de banda larga similares quelas empregadas em sistemas de cable
modem, e operava com cabo coaxial.
1BASE5 -- Uma tentativa antiga de padronizar uma soluo de LAN de baixo custo. Opera a 1 Mbit/s
e foi um fracasso comercial.
10BASE5 (tambm chamado Thicknet) -- Especificao Ethernet de banda bsica de 10 Mbps, que
usa o padro (grosso) de cabo coaxial de banda de base de 50 ohms. Faz parte da especificao de
camada fsica de banda de base IEEE 802.3, tem um limite de distncia de 500 metros por segmento.
10BASE-T -- Opera com 4 fios (dois conjuntos de par tranado) num cabo de cat-3 ou cat-5.
Um hub ou switch fica no meio e tem uma porta para cada n da rede. Essa tambm a configurao
usada para a ethernet 100BASE-T e a Gigabit.
FOIRL -- Link de fibra tica entre repetidores. O padro original para ethernet sobre fibra.
10BASE-F -- um termo genrico para a nova famlia de padres de ethernet de 10 Mbit/s: 10BASEFL, 10BASE-FB e 10BASE-FP. Desses, s o 10BASE-FL est em uso comum (todos utilizando a fibra
ptica como meio fsico).
10BASE-FB -- Pretendia ser usada por backbones conectando um grande nmero de hubs ou
switches, agora est obsoleta.
10BASE-FP -- Uma rede passiva em estrela que no requer repetidores, nunca foi implementada.
100BASE-T -- Designao para qualquer dos trs padres para 100 Mbit/s ethernet sobre cabo de
par tranado.
100BASE-T4 -- 100 Mbit/s ethernet sobre cabeamento cat-3 (Usada em instalaes 10BASE-T).
Utiliza todos os quatro pares no cabo. Atualmente obsoleto, cabeamento cat-5 o padro. Limitado a HalfDuplex.
100 Mbit/s ethernet sobre cabeamento cat-3. Suporta full-duplex, e usa apenas dois pares. Seu
funcionamento equivalente ao 100BASE-TX, mas suporta cabeamento antigo.
100BASE-FX -- 100 Mbit/s ethernet sobre fibra ptica. Usando fibra tica multimodo 62,5 mcrons
tem o limite de 400 metros.
1000BASE-LX -- 1 Gbit/s sobre fibra. Otimizado para distncias maiores com fibra mono-modo.
1000BASE-CX -- Uma soluo para transportes curtos (at 25m) para rodar ethernet de 1 Gbit/s num
cabeamento especial de cobre. Antecede o 1000BASE-T, e agora obsoleto.
10GBASE-SR -- projetado para suportar distncias curtas sobre cabeamento de fibra multi-modo,
variando de 26m a 82m dependendo do tipo de cabo. Suporta tambm operao a 300m numa fibra
multi-modo de 2000 MHz.
10GBASE-LX4 -- usa multiplexao por diviso de comprimento de onda para suportar distncias
entre 240m e 300m em cabeamento multi-modo. Tambm suporta 10 km com fibra mono-modo.
10GBASE-SW, 10GBASE-LW e 10GBASE-EW. Essas variedades usam o WAN PHY, projetado para
interoperar com equipamentos OC-192 / STM-64 SONET/SDH. Eles correspondem camada fsica do
10GBASE-SR, 10GBASE-LR e 10GBASE-ER respectivamente, e da usam os mesmos tipos de fibra e
suportam as mesmas distncias. (No h um padro WAN PHY correspondendo ao 10GBASE-LX4.)
100BaseVG - Um rival precoce para a ethernet de 100 Mbit/s. Ele roda com cabeamento categoria 3.
Usa quatro pares. Um fracasso, comercialmente.
CHAOSnet
Virtual LAN
telecomunicao
Internet
X25
Referncias
1. Ir para:a b c d O Que Ethernet (em portugus) Trabalhos Feitos (16 de setembro de 2009). Pgina visitada
em 30 de setembro de 2012.
2. Ir para:a b c d Ethernet o que isso ? (em portugus). Pgina visitada em 30 de setembro de 2012.
Ethernet: Distributed Packet Switching for Local Computer Networks Cpia em HTML do artigo de
1996, de Robert M. Metcalfe e David R. Boggs, parte dos textos lssicos da ACM
(em ingls)http://www.erg.abdn.ac.uk/users/gorry/course/lan-pages/enet.html
(em ingls)http://www.datacottage.com/nch/eoperation.htm
I - STARLAN
StarLAN was the first implementation of 1 megabit per second (1Mbit/s) Ethernet over twisted pair wiring. It was
standardized by the standards association of the Institute of Electrical and Electronics Engineers (IEEE) as 802.3e in
1986, as the 1BASE5 version of Ethernet.
References:
10 Mbit/sec:
Speeds
Terabit/sec
IEEE 802.3 Physical layer Autonegotiation Power over Ethernet Ethernet Type Ethernet Alliance Flow control
General
Frames Jumbos
Historical
Applications
Industrial Carrier Audio First mile Data center Energy Efficiency Synchronous
Transceivers
Interfaces
II LATTISNET
LattisNet was a family of computer networking hardware and software products built and sold by SynOptics
Communications (also rebranded by Western Digital)[citation
needed]
3000 series of LattisHub network hubs. LattisNet was the first implementation of 10 Megabits per second local area
networking over unshielded twisted pairwiring in a star topology.[1]
Contents
1 Ethernet variants
2 Ethernet compatibility
3 References
4 External links
Ethernet variants
During the early 1980s most networks used coaxial cable as the primary form of premise cabling
in Ethernet implementations. In 1985 SynOptics shipped its first hub for fiber optics and shielded twisted pair.
[1]
SynOptics' co-founder, Engineer Ronald V. Schmidt, had experimented with a fiber-optic variant of Ethernet called
Fibernet II while working at Xerox PARC, where Ethernet had been invented. [2] In January 1987 SynOptics announced
intentions to manufacture equipment supporting 10 megabits/sec data transfer rates over unshielded twisted pair,
telephone wire.[3]
In August 1987 New York based LAN Systems, Inc. completed the equipment testing and praised SynOptics for
successfully deploying a 10Mbit/s network that supported workstations up to 330 feet from the wiring closet, because of
their careful control of EMI and RFI. [3] Novell reported that the LattisNet equipment performed better than RG-58U coaxial
cable.[4]
This same year HP proposed a study group be formed to look into standardizing Ethernet on telephone wires. [3]
[5]
SynOptics' investor, Menlo Ventures explained its position on joining the IEEE for standardization.
When we [Menlo Ventures] initially made the investment, SynOptics had a pate
patent rights and defend our exclusivity, we basically have to give up the idea of becomin
In 1990 the IEEE issued an Ethernet over twisted pair standard known for transmitting 10 Mbit/s, or 10BASE-T
(802.3i).
Ethernet compatibility
Of the SynOptics hubs, the 2500 series was only compatible with LattisNet twisted-pair Ethernet; the 1000 and
3000 series featured modules for LattisNet and standard 10BASE-T. In the 1000 series, the 505 modules are LattisNet
and the 508 modules are 10BASE-T.
References
1.
:a b Urs von Burg (2001). "The Battle Between Ethernet and Token Ring". The Triumph of Ethernet. Stanford University
Press. ISBN 0-8047-4094-1.
2.
- R. Schmidt, E. Rawson, R. Norton Jr., S. Jackson, M. Bailey, (November 1983). "Fibernet II: A Fiber Optic
Ethernet". IEEE Journal on Selected Areas in Communications 1 (5): 702.doi:10.1109/JSAC.1983.1145992.
3.
:a b c Paula Musich (August 3, 1987). "User lauds SynOptic system: LattisNet a success on PDS". Network World 4 (31).
pp. 2, 39. Retrieved June 10, 2011.
4.
- Eric Killorin (November 2, 1987). "LattisNet makes the grade in Novell benchmark tests" 4 (44). Network World. p. 19.
Retrieved June 11, 2011.
5.
- Understanding the Network A Practical Guide to Internetworking, page 131, accessdate-20 March 2011
8P8C plug
Ethernet over twisted pair technologies use twisted-pair cables for the physical layer of an Ethernet computer
network. Other Ethernet cable standards employ coaxial cable or optical fiber. Early versions developed in the 1980s
included StarLAN followed by 10BASE-T. By the 1990s, fast, inexpensive technologies began to emerge. Currently the
most popular are 100BASE-TX (fast Ethernet; 100 Mbit/s) and 1000BASE-T (gigabit Ethernet; 1 Gbit/s). These standards
all use 8P8C connectors.[note
inclusively;
thus
it
is
1]
possible
to
mix
different
generations
of
equipment.
Inclusive
capability
is
designated 10/100 or 10/100/1000- for connections that support such combinations. [1]:123 The cables usually have four
pairs of wires (though 10BASE-T and 100BASE-TX only require two of the pairs). The three standards support both fullduplex and half-duplex communication. High-grade twisted pair cabling can transport up to 10 Gbit/s Ethernet
(10GBASE-T).
Contents
1 History
2 Naming
3 Cabling
5 Variants
6 See also
7 Notes
8 References
9 Further reading
10 External links
History
The Institute of Electrical and Electronics Engineers (IEEE) standards association ratified several versions of the
technology. The first two early designs were StarLAN, standardized in 1986, at one megabit per second, [2] and LattisNet,
developed in January 1987, at 10 megabit per second. [3][4] Both were developed before the 10BASE-T standard
(published in 1990 as IEEE 802.3i), and both were not compatible with it.[5]
Using twisted pair cabling, in a star topology, for Ethernet addressed several weaknesses of the previous
standards:
Twisted pair cables could be used more generally and were already present in many office buildings, lowering
overall cost.
The centralized star topology was a more common approach to cabling than the bus in earlier standards and
easier to manage.
Using point-to-point links instead of a shared bus greatly simplified troubleshooting and was less prone to failure.
Exchanging cheap repeater hubs for more advanced switching hubs provided a viable upgrade path.
Mixing different speeds in a single network became possible with the arrival of Fast Ethernet.
Naming
The common names for the standards derive from aspects of the physical media. The leading number (10 in
10BASE-T) refers to
The T designates twisted pair cable, where the pair of wires for each signal is twisted together to reduce radio frequency
interference and crosstalk between pairs. Where there are several standards for the same transmission speed, they are
distinguished by a letter or digit following the T, such as TX.
Cabling
Ppai
r
Wwire
Color
tip
white/orange
ring
orange
tip
white/green
ring
blue
tip
white/blue
ring
green
tip
white/brown
ring
brown
Twisted-pair Ethernet standards are such that the majority of cables can be wired "straight-through" (pin 1 to pin
1, pin 2 to pin 2 and so on), but others may need to be wired in the " crossover" form (receive to transmit and transmit to
receive).
10BASE-T and 100BASE-TX only require two pairs to operate, located on pins 1 plus 2 and pins 3 plus 6. Since
10BASE-T and 100BASE-TX need only two pairs and Category 5 cable has four pairs, it is possible, but not standards
compliant, to run two network connections or use spare pairs for PoE (Power over Ethernet) (or a network connection
and two phone lines) over a Category 5 cable by using the normally unused pairs (pins 45, 78) in 10- and 100-Mbit/s
configurations. In practice, great care must be taken to separate these pairs as most 10/100-Mbit/s hubs, switches, and
PCs electrically terminate the unused pins.[citation needed] Moreover, 1000BASE-T requires all four pairs to operate, pins 1 and
2, 3 and 6 as well as 4 and 5, 7 and 8.
It is conventional to wire cables for 10- or 100-Mbit/s Ethernet to either the T568A or T568B standards. Since
these standards differ only in that they swap the positions of the two pairs used for transmitting and receiving (TX/RX), a
cable with T568A wiring at one end and T568B wiring at the other is referred to as a crossover cable. The terms used in
the explanations of the 568 standards, tip and ring, refer to older communication technologies, and equate to the positive
and negative parts of the connections.
A 10BASE-T or 100BASE-TX node such as a PC, with a connector called medium dependent
interfaces (MDI), transmits on pin 1 and 2 and receives on pin 3 and 6 to a network device using a "straight-through"
cable. In order for two network devices or two nodes to communicate with each other (such as a switch to another switch
or computer to computer) a crossover cable is often required at speeds of 10 or 100 Mbit/s. If available, connections can
be made with a straight-through cable by means of an MDI-X port, also known as an "internal crossover" or "embedded
crossover" connection. Hub and switch ports with such internal crossovers are usually labelled as such, with "uplink" or
"X". For example, 3Com usually labels their ports 1X, 2X, and so on. In some cases a button is provided to allow a port to
act as either a normal or an uplink port.
Many modern Ethernet host adapters can automatically detect another computer connected with a straightthrough cable and then automatically introduce the required crossover, if needed; if neither of the adapters has this
capability, then a crossover cable is required. Most newer switches have automatic crossover ("auto MDI-X" or "autouplink") on all ports, eliminating the uplink port and the MDI/MDI-X switch, and allowing all connections to be made with
straight-through cables. If both devices being connected support 1000BASE-T according to the standards, they will
connect regardless of the cable being used or how it is wired.
A 10BASE-T transmitter sends two differential voltages, +2.5 V or 2.5 V.
100BASE-TX follows the same wiring patterns as 10BASE-T, but is more sensitive to wire quality and length, due
to the higher bit rates.
A 100BASE-TX transmitter sends three differential voltages, +1 V, 0 V, or 1 V.[6]
1000BASE-T uses all four pairs bi-directionally and the standard includes auto MDI-X; however, implementation
is optional. With the way that 1000BASE-T implements signaling, how the cable is wired is immaterial in actual usage.
The standard on copper twisted pair is IEEE 802.3ab for Cat 5e UTP, or 4D-PAM5; four dimensions using PAM (pulse
amplitude modulation) with five voltages, 2 V, 1 V, 0 V, +1 V, and +2 V[7] While +2 V to 2 V voltage may appear at the
pins of the line driver, the voltage on the cable is nominally +1 V, +0.5 V, 0 V, 0.5 V and 1 V.[8]
100BASE-TX and 1000BASE-T were both designed to require a minimum of Category 5 cable and also specify a
maximum cable length of 100 meters. Category 5 cable has since been deprecated and new installations use Category
5e.
Unlike
earlier
Ethernet
standards
cable,
such
as 10BASE5 (thicknet)
and 10BASE2 (thinnet), 10BASE-T does not specify the exact type of wiring to be used, but instead specifies certain
characteristics that a cable must meet. This was done in anticipation of using 10BASE-T in existing twisted-pair wiring
systems that may not conform to any specified wiring standard. Some of the specified characteristics
areattenuation, characteristic impedance, timing jitter, propagation delay, and several types of noise. Cable testers are
widely available to check these parameters to determine if a cable can be used with 10BASE-T. These characteristics are
expected to be met by 100 meters of 24-gauge unshielded twisted-pair cable. However, with high quality cabling, cable
runs of 150 meters or longer are often obtained and are considered viable by most technicians familiar with the 10BASET specification.[citation needed]
Variants
Speed
[Mbit/s]
Distance
[m]
Name
100
(nominally)
StarLAN
10
100
(nominally)
LattisNet
10
100
10BASE-T
(nominally)[11]
100
100
100BASE-TX
1000
100
1000BASE-T
10 000
100
10GBASE-T
40 000
30
40GBASE-T
Standard
/ Year
802.3e
1986[9]
Description
(pre) 802.3i Runs over AT&T Premises Distribution System (PDS) wiring or
four wires (two twisted pairs) on telephone twisted pair
1987
or Category 3 cable.[3][10]
802.3i
1990
802.3u
1995
802.3ab
1999
802.3an
2006
802.3bq[12]
See also
Ethernet extender
IEEE 802.3
Network isolator
Twisted pair
Notes
1.
- The 8P8C modular connector is often called RJ45 after a telephone industry standard.
References
1.
- Charles E. Spurgeon (2000). Ethernet: the definitive guide. OReilly Media. ISBN 978-1-56592-660-8.
2.
- Urs von Burg (2001). The triumph of Ethernet: technological communities and the battle for the LAN standard. Stanford
University Press. pp. 175176, 255256. ISBN 978-0-8047-4095-1.
3.
^ Jump up to:a b Paula Musich (August 3, 1987). "User lauds SynOptic system: LattisNet a success on PDS". Network
World 4 (31). pp. 2, 39. Retrieved June 10, 2011.
4.
- W.C. Wise, Ph.D. (March 1989). "Yesterday, somebody asked me what I think about LattisNet. Here's what I told him in a
nutshell". CIO Magazine 2 (6). p. 13. Retrieved June 11, 2011.(Advertisement)
5.
- Network Maintenance and Troubleshooting Guide. Fluke Networks. 2002. p. B-4. ISBN 1-58713-800-X.
6.
- David A. Weston (2001). Electromagnetic Compatibility: principles and applications. CRC Press. pp. 240242. ISBN 0-82478889-3. Retrieved June 11, 2011.
7.
- Steve Prior. "1000BASE-T Duffer's Guide to Basics and Startup". Retrieved 2011-02-18.
8.
- Nick van Bavel, Phil Callahan and John Chiang (2004-10-25). "Voltage-mode line drivers save on power". Retrieved 2011-0218.
9.
- 802.3a,b,c, and e-1988 IEEE Standards for Local Area Networks: Supplements to Carrier Sense Multiple Access With
Collision
Detection
(CSMA/CD) Access
Method
and
Physical
1987. doi:10.1109/IEEESTD.1987.78883.
10. - Eric Killorin (November 2, 1987). "LattisNet makes the grade in Novell benchmark tests" 4(44). Network World. p. 19.
Retrieved March 18, 2011.
11. - IEEE Computer Society (2008-12-26), IEEE Std 802.3-2008 : 14.1.1.3 Twisted-pair media, IEEE
12. - "IEEE P802.3bq 40GBASE-T Task Force". IEEE 802.3.
IV -
10GIGABIT ETHERNET
In computer networking, 10 gigabit Ethernet (10GE or 10GbE or 10 GigE) refers to various technologies for
transmitting Ethernet frames at a rate of 10 gigabits per second (10109 or 10 billion bits per second), first defined by
the IEEE 802.3ae-2002 standard. Unlike previous Ethernet standards, 10 gigabit Ethernet defines only full duplex pointto-point links which are generally connected by network switches; shared-medium CSMA/CD operation has not been
carried over from the previous generations Ethernet standards. [1] Half duplex operation and hubs do not exist in 10GbE.
Like previous versions of Ethernet, 10GbE supports both copper and fiber cabling. However, due to its higher
bandwidth requirements, higher-grade copper cables are required: category 6a orClass F/Category 7 cables for links up
to 100m. The 10 gigabit Ethernet standard encompasses a number of different physical layer (PHY) standards. A
networking device may support different PHY types through pluggable PHY modules, such as those based on SFP+.[2] At
the time that the 10 gigabit Ethernet standard was developed, interest in 10GbE as a wide area network (WAN) transport
led to the introduction of a WAN PHY for 10GbE. The WAN PHY encapsulates Ethernet packets in SONET OC-192c
frames and operates at a slightly slower data-rate (9.95328 Gbit/s) than the local area network (LAN) PHY.
Router with 10 gigabit Ethernet ports and three physical layer module types
The adoption of 10 gigabit Ethernet has been more gradual than previous revisions of Ethernet: in 2007, one
million 10GbE ports were shipped, in 2009 two million ports were shipped, and in 2010 over three million ports were
shipped, [3][4] with an estimated nine million ports in 2011. [5] As of 2012, the price per port of 10 gigabit Ethernet relative to
its one gigabit counterpart still hindered more widespread adoption, even though the price per gigabit of bandwidth
enabled by 10 gigabit Ethernet was already about one-third compared to the bandwidth cost of its one gigabit
predecessor.[6][7]
Contents
1 Standards
3 Optical fiber
o
3.1 10GBASE-SR
3.2 10GBASE-LR
3.3 10GBASE-LRM
3.4 10GBASE-ER
3.5 10GBASE-ZR
3.6 10GBASE-LX4
3.7 10GBASE-PR
4 Copper
o
4.1 10GBASE-CX4
4.3 Backplane
4.3.1 10GBASE-KX4
4.3.2 10GBASE-KR
4.4 10GBASE-T
6 10GbE NICs
7 See also
9 External links
Standards
Standard
Year
Description
802.3ae
2002[8]
10 Gbit/s Ethernet over fiber for LAN (10GBASE-SR, 10GBASE-LR, 10GBASE-ER, 10GBASE-LX4) and WAN
(10GBASE-SW, 10GBASE-LW, 10GBASE-EW)
802.3ak
2004
802.3-2005
2005
802.3an
2006
802.3ap
2007
802.3aq
2006
802.3-2008
2008
A revision of base standard incorporating the 802.3an/ap/aq/as amendments, two corrigenda and errata. Link aggregation
moved to 802.1AX
802.3av
2009
802.3-2012
2012
Over the years the Institute of Electrical and Electronics Engineers (IEEE) 802.3 working group has published
several standards relating to 10GbE. These included: 802.3ae-2002 (fiber -SR, -LR, -ER and -LX4 PMDs), 802.3ak-2004
(-CX4 copper twin-ax InfiniBand type cable), 802.3an-2006 (10GBASE-T copper twisted pair), 802.3ap-2007 (copper
backplane -KR and -KX4 PMDs) and 802.3aq-2006 (fiber -LRM PMD with enhanced equalization). The 802.3ae-2002
and 802.3ak-2004 amendments were consolidated into the IEEE 802.3-2005 standard. IEEE 802.3-2005 and the other
amendments were consolidated into IEEE Std 802.3-2008.
To support different 10GbE physical layer standards, many interfaces consist of a standard socket into which
different PHY modules may be plugged. Physical layer modules are not specified in an official standards body but
by multi-source
agreements (MSAs)
that
can
be
negotiated
more
quickly.
Relevant
MSAs
for
10GbE
include XENPAK (and related X2 and XPAK), XFP and SFP+. When choosing a PHY module, a designer considers cost,
reach, media type, power consumption, and size (form factor). A single point-to-point link can have different MSA
pluggable formats on either end (e.g. XPAK and SFP+) as long as the 10GbE optical or copper port type (e.g. 10GBASESR) inside the pluggable is identical.
XENPAK was the first MSA for 10GE and had the largest form factor. X2 and XPAK were later competing
standards with smaller form factors. X2 and XPAK have not been as successful in the market as XENPAK. XFP came
after X2 and XPAK and it is also smaller.
The newest module standard is the enhanced small form-factor pluggable transceiver, generally called SFP+.
Based on the small form-factor pluggable transceiver (SFP) and developed by the ANSI T11 fibre channel group, it is
smaller still and lower power than XFP. SFP+ has become the most popular socket on 10GE systems. [9][10] SFP+ modules
do only optical to electrical conversion, no clock and data recovery, putting a higher burden on the host's channel
equalization. SFP+ modules share a common physical form factor with legacy SFP modules, allowing higher port density
than XFP and the re-use of existing designs for 24 or 48 ports in a 19" rack width blade.
Optical modules are connected to a host by either a XAUI, XFI or SFI interface. XENPAK, X2, and XPAK
modules use XAUI to connect to their hosts. XAUI (XGXS) uses a four-lane data channel and is specified in IEEE 802.3
Clause 48. XFP modules use a XFI interface and SFP+ modules use an SFI interface. XFI and SFI use a single lane data
channel and the 64b/66b encodingspecified in IEEE 802.3 Clause 49.
SFP+ modules can further be grouped into two types of host interfaces: linear or limiting. Limiting modules are
preferred except when using old fiber infrastructure which requires the use of the linear interface provided by 10GBASELRM modules.[11]
Interconnect
AKA
Defined
[citation
2011
Connector[12]
Medium
Media Type
Wavelength
Max range
X2, SFP
fiber
serial multi-mode
850 nm
100 m
Notes
identical to 10GBASE-SR except for
lower performance/reach
10GBASE-USR
10GBASE-SR
short reach
2002
XENPAK, X2,
SFP+
fiber
serial multi-mode
850 nm
400 m
10GBASE-LR
long reach
2002
XENPAK, X2,
XFP, SFP+
fiber
serial single-mode
1310 nm
10 km
10GBASE-ER
extended reach
2002
XENPAK, X2,
XFP, SFP+
fiber
serial single-mode
1550 nm
40 km
10GBASE-ZR
XENPAK, X2,
XFP, SFP+
fiber
serial single-mode
1550 nm
80 km
10GBASE-LX4
2002
XENPAK, X2,
SFP+
fiber
WDM multi-mode
or single-mode
1310 nm
300 m (multimode), 10 km
(single-mode)
2006
XENPAK, X2,
SFP+
fiber
serial multi-mode
1310 nm
220 m
15 m
15 m
10GBASE-LRM
10GBASE-CX4
DA, "10GBASE-CR"
10GBASE-T
needed]
2004
XENPAK, X2
copper
InfiniBand
4Xtwinaxial 8pair[13]
2006
SFP+
copper
twinaxial 2-pair
2006
8P8C
copper
category 6, 6a or 7
twisted pair
1m
802.3ap
2007
copper
PCB backplane
10GBASE-KR
802.3ap
2007
copper
PCB backplane
1m
fiber
Passive Optical
Network
1270 nm/15
77 nm
20 km
802.3av
2009
10GBASE-KX4
10GBASE-PR
10G EPON
Optical fiber
A Foundry Router with 10 gigabit Ethernet optical interfaces (XFP transceiver). The yellow cables are single-mode duplex fiber
optic connections.
There are two classifications for optical fiber: single-mode (SMF) and multimode (MMF).[14] In SMF light follows a
single path through the fiber while in MMF it takes multiple paths resulting in differential mode delay (DMD). SMF is used
for long distance communication and MMF is used for distances of less than 300 m. SMF has a narrower core (8.3 m)
which requires a more precise termination and connection method. MMF has a wider core (50 or 62.5 m). The
advantage of MMF is that it can be driven by a low cost Vertical-cavity surface-emitting laser (VCSEL) for short distances,
and multimode connectors are cheaper and easier to terminate reliably in the field. The advantage of SMF is that it can
work over longer distances.[15]
In the 802.3 standard reference is made to FDDI-grade MMF fiber. This has a 62.5 m core and a minimum
modal
bandwidth
of
160 MHz*km
at
850 nm.
It
was
originally
installed
in
the
early
1990s
for FDDI and 100BaseFX networks. The 802.3 standard also references ISO/IEC 11801 which specifies optical MMF
fiber types OM1, OM2, OM3 and OM4. OM1 has a 62.5 m core while the others have a 50 m core. At 850 nm the
minimum modal bandwidth of OM1 is 200 MHz*km, of OM2 500 MHz*km, of OM3 2000 MHz*km and of OM4
4700 MHz*km. FDDI-grade cable is now obsolete and new structured cabling installations use either OM3 or OM4
cabling. OM3 cable can carry 10GbE 300 meters using low cost 10GBASE-SR optics (OM4 can manage 400 meters) . [16]
[17]
To distinguish SMF from MMF cables, SMF cables are usually yellow, while MMF cables are orange (OM1 &
OM2) or aqua (OM3 & OM4). However, in fiber optics there is no agreed color for any specific optical speed or
technology with the exception being angular physical connector (APC), it being an agreed color of green. [18]
There are also active optical cables (AOC). These have the optical electronics already connected eliminating the
connectors between the cable and the optical module. They plug into standard optical module sockets. They are lower
cost than other optical solutions because the manufacturer can match the electronics to the required length and type of
cable.
10GBASE-SR
10GBASE-SR ("short range") is a port type for multi-mode fiber and uses 850 nm lasers. Its Physical Coding
Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 49 and its Physical Medium Dependent PMD in Clause 52. It
delivers serialized data at a line rate of 10.3125 Gbit/s.
Over obsolete FDDI-grade 62.5 micrometers multimode fiber cabling it has a maximum range of 26 meters, over
62.5 micrometers OM1 it has a range of 33 meters, over 50 micrometers OM2 a range of 82 meters, over OM3
300 meters and over OM4 400 meters.[17] [19] OM3 and OM4 are the preferred choices for structured optical cabling within
buildings. MMF has the advantage over SMF of having lower cost connectors because of its wider core.
The 10GBASE-SR transmitter is implemented with a VCSEL which is low cost and low power. OM3 and OM4
optical cabling is sometimes described as laser optimized because they have been designed to work with VCSELs.
10GBASE-SR delivers the lowest cost, lowest power and smallest form factor optical modules.
For 2011, 10GBASE-SR is projected to make up a quarter of the total 10GbE adapter ports shipped. [20]
There is a non-standard lower cost, lower power variant sometimes referred to as 10GBASE-SRL (10GBASE-SR
lite). This is inter-operable with 10GBASE-SR but only has a reach of 100 meters.
10GBASE-LR
10GBASE-LR ("long reach") is a port type for single-mode fiber and uses 1310 nm lasers. Its Physical Coding
Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 49 and its Physical Medium Dependent PMD in Clause 52. It
delivers serialized data at a line rate of 10.3125 Gbit/s.
10GBASE-LR has a specified reach of 10 kilometres (6.2 mi), but 10GBASE-LR optical modules can often
manage distances of up to 25 kilometres (16 mi) with no data loss.
The 10GBASE-LR transmitter is implemented with a FabryProt or Distributed feedback laser (DFB). DFB
lasers are more expensive than VCSELs but their high power and longer wavelength allow efficient coupling into the
small core of single mode fiber over greater distances.
10GBASE-LRM
10GBASE-LRM, (Long Reach Multimode) originally specified in IEEE 802.3aq is a port type for multimode fiber
and uses 1310 nm lasers. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 49 and its Physical
Medium Dependent PMD in Clause 68. It delivers serialized data at a line rate of 10.3125 Gbit/s.
10GBASE-LRM supports distances up to 220 metres (720 ft) on FDDI-grade multimode fiber and the same 220m
maximum reach on OM1, OM2 and OM3 fiber types. [17] 10GBASE-LRM reach is not quite as far as the older 10GBASELX4 standard.
To ensure that specifications are met over FDDI-grade, OM1 and OM2 fibers, the transmitter should be coupled
through a mode conditioning patch cord. No mode conditioning patch cord is required for applications over OM3 or OM4.
[21]
Some 10GBASE-LRM transceivers also support distances up to 300 metres (980 ft) on standard single-mode
fiber (SMF, G.652), however this is not part of the IEEE or MSA specification.
10GBASE-LRM uses electronic dispersion compensation (EDC) for receive equalization. [22]
10GBASE-ER
10GBASE-ER ("extended reach") is a port type for single-mode fiber and uses 1550 nm lasers. Its Physical
Coding Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 49 and its Physical Medium Dependent PMD in Clause
52. It delivers serialized data at a line rate of 10.3125 Gbit/s.
The 10GBASE-ER transmitter is implemented with an externally modulated laser (EML).
10GBASE-ER has a reach of 40 kilometres (25 mi) over engineered links and 30 km over standard links.[17][23]
10GBASE-ZR
Several manufacturers have introduced 80 km (50 mi) range ER pluggable interfaces under the name
10GBASE-ZR. This 80 km PHY is not specified within the IEEE 802.3ae standard and manufacturers have created their
own specifications based upon the 80 km PHY described in the OC-192/STM-64 SDH/SONET specifications.
The 802.3 standard will not be amended to cover the ZR PHY.
10GBASE-LX4
10GBASE-LX4 is a port type for multimode fiber and single-mode fiber. It uses four separate laser sources
operating at 3.125 Gbit/s and coarse WDM with four unique wavelengths around 1310 nm. Its Physical Coding
Sublayer 8B10B PCS is defined in IEEE 802.3 Clause 48 and its Physical Medium Dependent PMD in Clause 53.[17]
It supports a range of 300 metres (980 ft) over FDDI-grade, OM1, OM2 and OM3 multimode cabling (all these
fiber types are specified to have a minimum modal bandwidth of 500 MHz*km at 1300 nm).
10GBASE-LX4 also supports a range of 10 kilometres (6.2 mi) over SMF.
For MMF links the WDM output needs to be coupled through a SMF offset-launch mode-conditioning patch cord.
This is explained in subclauses 53.6 and 38.11.4 of the IEEE 802.3 spec. [17]
Until 2005 10GBASE-LX4 optical modules were cheaper than 10GBASE-LR optical modules.
10GBASE-LX4 was used by people who wanted to support both MMF and SMF with a single optical module.
10GBASE-LX4 is now an obsolete technology and has no significant market presence.
10GBASE-PR
10GBASE-PR ("PON") originally specified in IEEE 802.3av is a 10G Ethernet PHY for passive optical networks
and uses 1577 nm lasers in the down stream direction and 1270 nm lasers in the upstream direction. Its Physical Medium
Dependent PMD is specified in Clause 75. Downstream it delivers serialized data at a line rate of 10.3125 Gbit/s in a
point to multi-point configuration.[17]
10GBASE-PR has three power budgets specified as 10GBASE-PR10, 10GBASE-PR20 and 10GBASE-PR30.
Copper
10G Ethernet can also run over twin-axial cabling, twisted pair cabling, and backplanes.
10GBASE-CX4
10GBASE-CX4 was the first 10G copper standard published by 802.3 (as 802.3ak-2004). It uses the XAUI 4lane PCS (Clause 48) and copper cabling similar to that used by InfiniBand technology. It is specified to work up to a
distance of 15 m (49 ft). Each lane carries 3.125 G baud of signaling bandwidth.
10GBASE-CX4 offers the advantages of low power, low cost and low latency, but has a bigger form factor and
more bulky cables than the newer single lane SFP+ standard and a much shorter reach than fiber or 10GBASE-T. This
cable is fairly rigid and considerably more costly than Category 5 or 6 UTP.
Shipments of 10GBASE-CX4 today are very low. [20] although some network vendors offer CX-4 interfaces which
can be used for either 10GBase ethernet or for stacking of switches at (slightly) higher speeds. An example of combi
stacking/ethernet are Dell PowerConnect PCT6200, PCT7000 and the 1G Powerconnect blade switches PCM6220 and
PCM6348[24]
Backplane
Backplane Ethernet also known by its task force name 802.3ap is used in backplane applications such
as blade servers and modular routers/switches with upgradable line cards. 802.3ap implementations are required to
operate in an environment comprising up to 1 metre (39 in) of copper printed circuit board with two connectors. The
standard defines two port types for 10 Gbit/s (10GBASE-KX4 and 10GBASE-KR) and a 1 Gbit/s port type (1000BASEKX). It also defines an optional layer for FEC, a backplane autonegotiation protocol and link training for 10GBASE-KR
where the receiver can set a three tap transmit equalizer. The autonegotiation protocol selects between 1000BASE-KX,
10GBASE-KX4, 10GBASE-KR or 40GBASE-KR4 operation. 40GBASE-KR4 is defined in 802.3ba.
[28]
10GBASE-KX4
This operates over four backplane lanes and uses the same physical layer coding (defined in IEEE 802.3 Clause
48) as 10GBASE-CX4.
10GBASE-KR
This operates over a single backplane lane and uses the same physical layer coding (defined in IEEE 802.3
Clause 49) as 10GBASE-LR/ER/SR.
10GBASE-T
10GBASE-T, or IEEE 802.3an-2006, is a standard released in 2006 to provide 10 Gbit/s connections over
unshielded or shielded twisted pair cables, over distances up to 100 metres (330 ft).[29]Category 6a is required to reach
the full distance of 100 metres (330 ft) and category 6 may reach a distance of 55 metres (180 ft) depending on the
quality of installation, determined only after re-testing to 500MHz. 10GBASE-T cable infrastructure can also be used for
1000BASE-T allowing a gradual upgrade from 1000BASE-T using autonegotiation to select which speed to use.
10GBASE-T has latency in the range 2 to 4 microseconds compared to 1 to 12 microseconds on 1000BASE-T. [30][31] As of
2010 10GBASE-T silicon is available from several manufacturers [32][33][34][35] with claimed power dissipation of 34 W at
structure widths of 40 nm, and with 28 nm in development, power will continue to decline.[36]
10GBASE-T uses the IEC 60603-7 8P8C (commonly known as RJ45) connectors already widely used with
Ethernet. Transmission characteristics are now specified to 500 MHz. To reach this frequency Category 6A or better
balanced twisted pair cables specified in ISO/IEC 11801 amendment 2 or ANSI/TIA-568-C.2 are needed to carry
10GBASE-T up to distances of 100 m. Category 6 cables can carry 10GBASE-T for shorter distances when qualified
according to the guidelines in ISO TR 24750 or TIA-155-A.
The 802.3an standard specifies the wire-level modulation for 10GBASE-T to be Tomlinson-Harashima
precoded (THP) pulse-amplitude modulation with 16 discrete levels (PAM-16), encoded in a two-dimensional
checkerboard pattern known as DSQ128. [37] Prior to precoding, forward error correction (FEC) coding is performed using
a (2048,172) low-density parity-check code, with the parity check matrix construction based on a generalized Reed
Solomon (32,2,31) code over GF(26).[37] By contrast PAM-5 is the modulation technique used in 1000BASE-T gigabit
Ethernet.
10GbE NICs
10GbE network interface cards are available from several manufacturers. These plug into ordinary computer
servers using PCI express and provide one or more PHY module, LC or 8P8Cconnectors.
See also
100 Gigabit Ethernet
GG45
TERA
XAUI
Optical interconnect
Optical communication
10G-PON
- Michael Palmer. Hands-On Networking Fundamentals, 2nd ed.. Cengage Learning. p. 180. ISBN 978-1-285-40275-8.
2.
- Anil Sharma (19 January 2011). "LightCounting forecasts CAGR of Over 300 Percent for 10GBASE-T Port Shipments
Through 2014". TMCnet. Retrieved 7 May 2011.
3.
4.
5.
- http://www.wired.com/wiredenterprise/2012/03/google-microsoft-network-gear/
6.
7.
8.
9.
10. - "10GbE Optical Component and SFP+ Modules: This Time It's Different by Andrew Schmitt". Retrieved 11 March 2008.
11. - "The road to SFP+: Examining module and system architectures by Ryan Latchman and Bharat Tailor". Retrieved 15
January 2009.
12. - 10-Gigabit Ethernet Transceiver Modules Compatibility Matrix
13. - Practical deployment and management of InfiniBand
14. - "Optical Fiber and 10 gigabit Ethernet white paper by the 10GEA". Archived from the original on 14 June 2008.
Retrieved 1 July 2008.
15. - "Why choose Multimode fiber? by Corning". Retrieved 11 March 2008.
16. - "10 Gigabit Ethernet over Multimode Fiber by John George". Retrieved 10 March 2008.
17. ^ Jump up to:a b c d e f g h "IEEE 802.3 standard".
18. - "How to tell? MMF or SMF". Retrieved 6 September 2011.
19. - "Description of Cisco 10G optical modules". Retrieved 3 May 2010.
20. ^ Jump up to:a b c d "Another Serving of Alphabet Soup by Intel". Retrieved 4 September 2011.
21. - "Cisco 10GBASE SFP+ Modules Data Sheet". Cisco Systems. February 2012. Retrieved 2012-05-12.
22. - "10GBase-LX4 vs 10GBase-LRM: A debate". Archivedfrom the original on 2009-07-20. Retrieved 2009-07-16.
23. - "Cisco 10GBASE XENPAK Modules". Cisco Systems. November 2011. Retrieved 2012-05-12.
24. - Dell webshop CX4 uplink and stacking module, visited 2 March 2013
25. - "Cables and Transceivers". Arista Networks. Retrieved 21 September 2012.
26. - "Optcore SFP+ direct-attach cables". Optcore. Retrieved 21 September 2012.
27. - "HP X242 SFP+ Direct Attach Copper Cable". Hewlett Packard. Retrieved 27 March 2013.
28. - "IEEE P802.3ap Backplane Ethernet Task Force". Retrieved 30 January 2011.
29. - "IEEE Standards Status Report for 802.3an". Retrieved 14 August 2007.
30. - 10GBASE-T for Broad 10 Gigabit Adoption in the Data Center, Intel, retrieved 2011-12-21
31. - SWITCHES SWITCH FROM 1000BASET TO 10GBASET NOW, Teranetics, October 2009, retrieved 2011-12-21
32. - "Broadcom 10GBASE-T PHY". Retrieved 2 December 2011.
33. - "PLX Technology, Teranetics 10GBASE-T PHY". Retrieved 11 February 2011.
34. - "Solar Flare 10GBASE-T PHY". Archived from the original on 2009-09-07. Retrieved 2009-09-05.
35. - "Aquantia 10GBASE-T PHY". Archived from the original on 3 December 2008. Retrieved 10 December 2008.
36. - Hostetler, Jeff. "10GBASE-T Is 2012 the Year for Wide Adoption?".
37.
- Ungerboeck, Gottfried (22 September 2006)."10GBASE-T: 10Gbit/s Ethernet over copper". Vienna: Broadcom.
Retrieved 7 August 2013.
In computer networking, Fast Ethernet is a collective term for a number of Ethernet standards that carry traffic at
the nominal rate of 100 Mbit/s, against the original Ethernet speed of 10 Mbit/s. Of the Fast Ethernet standards
100BASE-TX is by far the most common and is supported by the vast majority of Ethernet hardware currently produced.
Fast Ethernet was introduced in 1995[1] and remained the fastest version of Ethernet for three years before being
superseded by gigabit Ethernet.[2]
Contents
1 General design
2 Copper
o
2.1 100BASE-TX
2.2 100BASE-T4
2.3 100BASE-T2
3 Fiber optics
o
3.1 100BASE-FX
3.2 100BASE-SX
3.3 100BASE-BX
3.4 100BASE-LX10
4 See also
5 References
6 External links
General design
Fast Ethernet is an extension of the existing Ethernet standard. It runs on UTP data or optical fiber cable in a star
wired bus topology, similar to 10BASE-T where all cables are attached to a hub. And, it provides compatibility with
existing 10BASE-T systems and thus enables plug-and-play upgrades from 10BASE-T. Fast Ethernet is sometimes
referred to as 100BASE-X where X is a placeholder for the FX and TX variants.[citation needed] The standard specifies the use
of CSMA/CD for media access control, although in practice all modern networks use Ethernet switches and operate
in full-duplex mode.
The 100 in the media type designation refers to the transmission speed of 100 Mbit/s. The "BASE" refers
to baseband signalling. The TX, FX and T4 refer to the physical medium that carries the signal.
A Fast Ethernet adapter can be logically divided into a Media Access Controller (MAC) which deals with the
higher level issues of medium availability and a Physical Layer Interface (PHY). The MAC may be linked to the PHY by a
4 bit 25 MHz synchronous parallel interface known as a Media Independent Interface (MII) or a 2 bit 50 MHz
variant Reduced Media Independent Interface(RMII). Repeaters (hubs) are also allowed and connect to multiple PHYs for
their different interfaces.
The MII may (rarely) be an external connection but is usually a connection between ICs in a network adapter or
even within a single IC. The specs are written based on the assumption that the interface between MAC and PHY will be
a MII but they do not require it.
The MII fixes the theoretical maximum data bit rate for all versions of Fast Ethernet to 100 Mbit/s. The data
signaling rate actually observed on real networks is less than the theoretical maximum, due to the necessary header and
trailer (addressing and error-detection bits) on every frame, the occasional "lost frame" due to noise, and time waiting
after each sent frame for other devices on the network to finish transmitting.
Copper
100BASE-T is any of several Fast Ethernet standards for twisted pair cables, including: 100BASE-TX
(100 Mbit/s over two-pair Cat5 or better cable), 100BASE-T4 (100 Mbit/s over four-pair Cat3 or better cable, defunct),
100BASE-T2 (100 Mbit/s over two-pair Cat3 or better cable, also defunct). The segment length for a 100BASE-T cable is
limited to 100 metres (328 ft) (as with 10BASE-T and gigabit Ethernet). All are or were standards under IEEE
802.3 (approved 1995). Almost all 100BASE-T installations are 100BASE-TX.
In the early days of Fast Ethernet, much vendor advertising centered on claims by competing standards that said
vendors' standards will work better with existing cables than other standards. In practice, it was quickly discovered that
few existing networks actually met the assumed standards, because 10-megabit Ethernet was very tolerant of minor
deviations from specified electrical characteristics and few installers ever bothered to make exact measurements of cable
and connection quality; if Ethernet worked over a cable, no matter how well it worked, it was deemed acceptable. Thus
most networks had to be rewired for 100-megabit speed whether or not there had supposedly been CAT3 or CAT5 cable
runs.[citation needed]
100BASE-TX[edit]
8P8C Wiring (TIA/EIA-568-B T568B)
Pin
Pair
Wire
Color
+/tip
white/orange
/ring
orange
+/tip
white/green
/ring
blue
+/tip
white/blue
/ring
green
+/tip
white/brown
/ring
brown
100BASE-TX is the predominant form of Fast Ethernet, and runs over two wire-pairs inside a category 5 or above
cable. Like 10BASE-T, the active pairs in a standard connection are terminated on pins 1, 2, 3 and 6. Since a typical
category 5 cable contains 4 pairs, it can support two 100BASE-TX links with a wiring adaptor. [3] Cabling is conventional
wired to TIA/EIA-568-B's termination standards, T568A or T568B. This places the active pairs on the orange and green
pairs (canonical second and third pairs).
Each network segment can have a maximum cabling distance of 100 metres (328 ft). In its typical configuration,
100BASE-TX uses one pair of twisted wires in each direction, providing 100 Mbit/s of throughput in each direction (fullduplex). See IEEE 802.3 for more details.
The configuration of 100BASE-TX networks is very similar to 10BASE-T. When used to build a local area
network, the devices on the network (computers, printers etc.) are typically connected to a hub or switch, creating a star
network. Alternatively it is possible to connect two devices directly using a crossover cable.
With 100BASE-TX hardware, the raw bits (4 bits wide clocked at 25 MHz at the MII) go through 4B5B binary
encoding to generate a series of 0 and 1 symbols clocked at 125 MHz symbol rate. The 4B5B encoding provides DC
equalization and spectrum shaping (see the standard for details). Just as in the 100BASE-FX case, the bits are then
transferred to the physical medium attachment layer using NRZI encoding. However, 100BASE-TX introduces an
additional, medium dependent sublayer, which employs MLT-3 as a final encoding of the data stream before
transmission, resulting in a maximum "fundamental frequency" of 31.25 MHz. The procedure is borrowed from the ANSI
X3.263 FDDI specifications, with minor discrepancies.[4]
100BASE-T4
100BASE-T4 was an early implementation of Fast Ethernet. It requires four twisted copper pairs, but those pairs
were only required to be category 3 rather than the category 5 required by TX. One pair is reserved for transmit, one for
receive, and the remaining two will switch direction as negotiated. A very unusual 8B6T code is used to convert 8 data
bits into 6 base-3 digits (the signal shaping is possible as there are nearly three times as many 6-digit base-3 numbers as
there are 8-digit base-2 numbers). The two resulting 3-digit base-3 symbols are sent in parallel over 3 pairs using 3level pulse-amplitude modulation (PAM-3). The fact that 3 pairs are used to transmit in each direction makes 100BASET4 inherently half-duplex. This standard can be implemented with CAT 3, 4, 5 UTP cables, or STP if needed against
interference. Maximum distance is limited to 100 meters. 100BASE-T4 was not widely adopted but the technology
developed for it is used in 1000BASE-T.[5]
100BASE-T2
Symbol
000
001
+1
010
011
100 (ESC)
+2
In 100BASE-T2, the data is transmitted over two copper pairs, 4 bits per symbol. It uses these two pairs for
simultaneously transmitting and receiving on both pairs [6] thus allowing full-duplex operation. First, a 4 bit symbol is
expanded into two 3-bit symbols through a non-trivial scrambling procedure based on a linear feedback shift register; see
the standard for details. This is needed to flatten the bandwidth and emission spectrum of the signal, as well as to match
transmission line properties. The mapping of the original bits to the symbol codes is not constant in time and has a fairly
large period (appearing as a pseudo-random sequence). The final mapping from symbols to PAM-5 line modulation levels
obeys the table on the right. 100BASE-T2 was not widely adopted but the technology developed for it is used in
1000BASE-T.[5]
Fiber optics
100BASE-FX
100BASE-FX is a version of Fast Ethernet over optical fiber. It uses a 1300 nm near-infrared (NIR)
light wavelength transmitted via two strands of optical fiber, one for receive(RX) and the other for transmit(TX). Maximum
length is 412 metres (1,350 ft)[citation needed] for half-duplex connections (to ensure collisions are detected), and 2 kilometres
(6,600 ft) for full-duplex over multi-mode optical fiber. 100BASE-FX uses the same 4B5B encoding and NRZI line code
that 100BASE-TX does. 100BASE-FX should use SC, ST, LC, MTRJ or MIC connectors with SC being the preferred
option.[7]
100BASE-FX is not compatible with 10BASE-FL, the 10 MBit/s version over optical fiber.
100BASE-SX
100BASE-SX is a version of Fast Ethernet over optical fiber. It uses two strands of multi-mode optical fiber for
receive and transmit. It is a lower cost alternative to using 100BASE-FX, because it uses short wavelength optics which
are significantly less expensive than the long wavelength optics used in 100BASE-FX. 100BASE-SX can operate at
distances up to 550 metres (1,800 ft).
100BASE-SX uses the same wavelength as 10BASE-FL, the 10 Mbit/s version over optical fiber. Unlike
100BASE-FX, this allows 100BASE-SX to be backwards-compatible with 10BASE-FL.
Because of the shorter wavelength used (850 nm) and the shorter distance it can support, 100BASE-SX uses
less expensive optical components (LEDs instead of lasers) which makes it an attractive option for those upgrading from
10BASE-FL and those who do not require long distances.
100BASE-SX is not standardized by the IEEE 802.3 committee. It is an industry de facto standard rather than a
formal Ethernet standard.[citation needed]
100BASE-BX
100BASE-BX is a version of Fast Ethernet over a single strand of optical fiber (unlike 100BASE-FX, which uses
a pair of fibers). Single-mode fiber is used, along with a special multiplexer which splits the signal into transmit and
receive wavelengths; the two wavelengths used for transmit and receive are 1310 nm and 1550 nm. The terminals on
each side of the fiber are not equal, as the one transmitting "downstream" (from the center of the network to the outside)
uses the 1550 nm wavelength, and the one transmitting "upstream" uses the 1310 nm wavelength. Distances can be 10,
20 or 40 km.
100BASE-LX10
100BASE-LX10 is a version of Fast Ethernet over two single-mode optical fibers. It has a nominal reach of
10 km and a nominal wavelength of 1310 nm. It is described in IEEE 802.3-2005 Section 5 chapter 58
2.
3.
- "The 100BASE-TX PMD (and MDI) is specified by incorporating the FDDI TP-PMD standard, ANSI X3.263: 1995 (TPPMD), by reference, with the modifications noted below." (section 25.2 of IEEE802.3-2002).
4.
- Charles E. Spurgeon (2000). Ethernet: the Definitive Guide. O'Reilly Media. p. 156. ISBN 978-1-56592-660-8.
5.
- Robert Breyer and Sean Riley (1999). Switched, Fast, and Gigabit Ethernet. Macmillan Technical Publishing. p. 107.
6.
VI - GIGABIT ETHERNET
( 1000 BASE-T / 1GbBASE-T )
GigE redirects here. For the camera protocol, see GigE vision.
In computer networking, gigabit Ethernet (GbE or 1 GigE) is a term describing various technologies for
transmitting Ethernet frames at a rate of a gigabit per second (1,000,000,000 bits per second), as defined by
the IEEE 802.3-2008 standard. It came into use beginning in 1999, gradually supplanting Fast Ethernet in wired local
networks, where it performed considerably faster. The cables and equipment are very similar to previous standards
and have been very common and economical since 2010.
Half-duplex gigabit links connected through hubs are allowed by the specification,[1] but full-duplex usage
with switches is used exclusively.
Contents
1 History
2 Varieties
2.1 1000BASE-X
2.1.1 1000BASE-CX
2.1.2 1000BASE-KX
2.1.3 1000BASE-SX
2.1.4 1000BASE-LX
2.1.5 1000BASE-LX10
2.1.6 1000BASE-EX
2.1.7 1000BASE-BX10
2.1.8 1000BASE-ZX
2.2 1000BASE-T
2.3 1000BASE-TX
3 See also
4 Notes
5 References
6 Further reading
7 External links
History
Ethernet was the result of the research done at Xerox PARC in the early 1970s. Ethernet later evolved into a
widely implemented physical and link layer protocol. Fast Ethernet increased speed from 10 to 100 megabits per second
(Mbit/s). Gigabit Ethernet was the next iteration, increasing the speed to 1000 Mbit/s. The initial standard for gigabit
Ethernet was produced by the IEEE in June 1998 as IEEE 802.3z, and required optical fiber. 802.3z is commonly
referred to as 1000BASE-X, where -X refers to either -CX, -SX, -LX, or (non-standard) -ZX. For the history behind the "X"
see Fast Ethernet.
IEEE
802.3ab,
ratified
in
1999,
defines
gigabit
Ethernet
transmission
pair (UTP) category 5, 5e, or 6 cabling and became known as 1000BASE-T. With the ratification of 802.3ab, gigabit
Ethernet became a desktop technology as organizations could use their existing copper cabling infrastructure.
IEEE 802.3ah, ratified in 2004 added two more gigabit fiber standards, 1000BASE-LX10 (which was already
widely implemented as vendor specific extension) and 1000BASE-BX10. This was part of a larger group of protocols
known as Ethernet in the First Mile.
Initially, gigabit Ethernet was deployed in high-capacity backbone network links (for instance, on a high-capacity
campus network). In 2000, Apple's Power Mac G4 and PowerBook G4 were the first mass-produced personal computers
featuring the 1000BASE-T connection.[2] It quickly became a built-in feature in many other computers.
Higher bandwidth 10 Gigabit Ethernet standards have since become available as the IEEE ratified a fiber-based
standard in 2002, and a twisted pair standard in 2006. As of 2009, 10Gb Ethernet is replacing 1Gb as the backbone
network and has begun to migrate down to high-end server systems. [citation needed]
Varieties
1000BASE-T capable network interface card made by Intel, which connects to the computer via PCI-X
There are five physical layer standards for gigabit Ethernet using optical fiber (1000BASE-X), twisted pair
cable (1000BASE-T), or shielded balanced copper cable (1000BASE-CX).
The IEEE 802.3z standard includes 1000BASE-SX for transmission over multi-mode fiber, 1000BASE-LX for
transmission over single-mode fiber, and the nearly obsolete 1000BASE-CX for transmission over shielded balanced
copper cabling. These standards use 8b/10b encoding, which inflates the line rate by 25%, from 1000 Mbit/s to
1250 Mbit/s, to ensure a DC balanced signal. The symbols are then sent using NRZ.
IEEE 802.3ab, which defines the widely used 1000BASE-T interface type, uses a different encoding scheme in
order to keep the symbol rate as low as possible, allowing transmission over twisted pair.
IEEE 802.3ap defines Ethernet Operation over Electrical Backplanes at different speeds.
Ethernet in the First Mile later added 1000BASE-LX10 and -BX10.
Name
Medium
Specified distance
1000BASE-C
Shielded balanced copper cable[3]
X
25 meters
1000BASE-K
Copper backplane
X
1 meter
1000BASE-S
Multi-mode fiber
X
1000BASE-L
Multi-mode fiber
X
550 meters[5]
1000BASE-L
Single-mode fiber
X
5 km[5]
1000BASE-L
Single-mode fiber using 1,310 nm wavelength
X10
10 km[5]
1000BASE-E
Single-mode fiber at 1,310 nm wavelength
X
~ 40 km
1000BASE-Z
Single-mode fiber at 1,550 nm wavelength
X
~ 70 km
100 meters
1000BASE-T
Twisted-pair cabling (Cat-6, Cat-7)
X
100 meters
1000BASE-X
1000BASE-X is used in industry to refer to gigabit Ethernet transmission over fiber, where options include
1000BASE-SX, 1000BASE-LX, 1000BASE-LX10, 1000BASE-BX10 or the non-standard -EX and -ZX implementations.
1000BASE-CX
1000BASE-CX is an initial standard for gigabit Ethernet connections with maximum distances of 25 meters using
balanced shielded twisted pair and either DE-9 or 8P8C connector (with a pinout different from 1000BASE-T). The short
segment length is due to very high signal transmission rate. Although it is still used for specific applications where cabling
is done by IT professionals, for instance the IBM BladeCenter uses 1000BASE-CX for the Ethernet connections between
the blade servers and the switch modules, 1000BASE-T has succeeded it for general copper wiring use.
1000BASE-KX
1000BASE-KX is part of the IEEE 802.3ap standard for Ethernet Operation over Electrical Backplanes. This
standard defines 1-4 lanes of backplane links, one RX and one TX differential pair per lane, at link bandwidth ranging
from 100Mbit to 10Gbit per second (from 100BASE-KX to 10GBASE-KX4). The 1000BASE-KX variant uses 1.25 GBd
electrical (not optical) signalling speed.
1000BASE-SX
1000BASE-SX is a fiber optic gigabit Ethernet standard for operation over multi-mode fiber using a 770 to
860 nanometer, near infrared (NIR) light wavelength.
The standard specifies a distance capability between 220 metres (62.5/125 m fiber with low modal bandwidth)
and 550 metres (50/125 m fiber with high modal bandwidth). In practice, with good quality fiber, optics, and terminations,
1000BASE-SX will usually work over significantly longer distances. [citation needed]
This standard is highly popular for intra-building links in large office buildings, co-location facilities and carrier
neutral internet exchanges.
Optical power specifications of SX interface: Minimum output power = 9.5 dBm. Minimum receive sensitivity =
17 dBm.
1000BASE-LX
1000BASE-LX is a fiber optic gigabit Ethernet standard specified in IEEE 802.3 Clause 38 which uses a long
wavelength laser (1,2701,355 nm), and a maximum RMS spectral width of 4 nm.
1000BASE-LX is specified to work over a distance of up to 5 km over 10 m single-mode fiber.
1000BASE-LX can also run over all common types of multi-mode fiber with a maximum segment length of 550
m. For link distances greater than 300 m, the use of a special launch conditioning patch cord may be required. [6] This
launches the laser at a precise offset from the center of the fiber which causes it to spread across the diameter of the
fiber core, reducing the effect known as differential mode delay which occurs when the laser couples onto only a small
number of available modes in multi-mode fiber.
1000BASE-LX10
1000BASE-LX10 was standardized six years after the initial gigabit fiber versions as part of the Ethernet in the
First Mile task group. It is very similar to 1000BASE-LX, but achieves longer distances up to 10 km over a pair of singlemode fiber due to higher quality optics. Before it was standardized 1000BASE-LX10 was essentially already in
widespread use by many vendors as a proprietary extension called either 1000BASE-LX/LH or 1000BASE-LH. [7]
1000BASE-EX
1000BASE-EX is a non-standard but industry accepted [citation needed] term to refer to gigabit Ethernet transmission. It
is very similar to 1000BASE-LX10 but achieves longer distances up to 40 km over a pair of single-mode fibers due to
higher quality optics than a LX10, running on 1310 nm wavelength lasers.[8] It is sometimes referred to as LH (Long Haul).
Easily confused with a 1000BASE-LX10 or 1000BASE-ZX because some vendors use the LH term.
1000BASE-BX10
1000BASE-BX10 is capable of up to 10 km over a single strand of single-mode fiber, with a different wavelength
going in each direction. The terminals on each side of the fibre are not equal, as the one transmitting downstream (from
the center of the network to the outside) uses the 1,490 nm wavelength, and the one transmitting upstream uses the
1,310 nm wavelength.
1000BASE-ZX
1000BASE-ZX is a non-standard but industry accepted [citation needed] term to refer to gigabit Ethernet transmission
using 1,550 nm wavelength to achieve distances of at least 70 km over single-mode fiber.
1000BASE-T
1000BASE-T (also known as IEEE 802.3ab) is a standard for gigabit Ethernet over copper wiring.
Each 1000BASE-T network segment can be a maximum length of 100 meters (328 feet), and must use Category
5 cable or better (including Cat 5e andCat 6).
Autonegotiation is a requirement for using 1000BASE-T [9] according to Section 28D.5 Extensions required for
Clause40 (1000BASE-T).[10] At least the clock source has to be negotiated, as one endpoint must be master and the
other endpoint must be slave.
In a departure from both 10BASE-T and 100BASE-TX, 1000BASE-T uses all four cable pairs for simultaneous
transmission in both directions through the use of adaptive equalization and a 5-level pulse amplitude modulation (PAM5) technique. The symbol rate is identical to that of 100BASE-TX (125Mbaud) and the noise immunity of the 5-level
signaling is also identical to that of the 3-level signaling in 100BASE-TX, since 1000BASE-T uses 4-dimensional trellis
coded modulation (TCM) to achieve a 6 dB coding gain across the 4 pairs.
Since negotiation takes place on only two pairs, if two gigabit devices are connected through a cable with only
two pairs, the devices will successfully choose 'gigabit' as the highest common denominator (HCD), but the link will never
come up. Most gigabit physical devices have a specific register to diagnose this behaviour. Some drivers offer an
"Ethernet@Wirespeed" option where this situation leads to a slower yet functional connection. [11]
The data is transmitted over four copper pairs, eight bits at a time. First, eight bits of data are expanded into four
3-bit symbols through a non-trivial scrambling procedure based on a linear feedback shift register; this is similar to what is
done in 100BASE-T2, but uses different parameters. The 3-bit symbols are then mapped to voltage levels which vary
continuously during transmission. An example mapping is as follows:
Symbol
000
001
010
011
100
101
110
111
+1
+2
+1
Automatic MDI/MDI-X Configuration is specified as an optional feature in the 1000BASE-T standard, [12] meaning
that straight-through cables will often work between gigabit-capable interfaces. This feature eliminates the need
for crossover cables, making obsolete the uplink/normal ports and manual selector switches found on many older hubs
and switches and greatly reduces installation errors.
1000BASE-TX[edit]
The Telecommunications Industry Association (TIA) created and promoted a standard similar to 1000BASE-T
that was simpler to implement, calling it 1000BASE-TX (TIA/EIA-854). [13] The simplified design would have, in theory,
reduced the cost of the required electronics by only using two unidirectional pairs in each direction instead of 4
bidirectional. However, this solution has been a commercial failure, likely due to the required Category 6 cabling and the
rapidly falling cost of 1000BASE-T products.
1000BASE-T products are sometimes marketed as 1000BASE-TX despite the difference in standards. The
confusion possibly stems from the most popular form of Fast Ethernet (100 Mbit/s) is known as 100BASE-TX, leading to
many products supporting multiple speeds of 10/100/1000Mbit/s marketed as "10/100/1000BASE-TX". [note 1]
See also
List of device bandwidths
Jumbo frames
Notes
- An example of a product specifying 10/100/1000BASE-TX ports can be found at "Cisco SR224G 24-port 10/100 2port Gigabit Switch + 2 miniGBIC". Archived from the original on September 10, 2011.
References
- A single repeater per collision domain is defined in IEEE 802.3 2008 Section 3: 41. Repeater for 1000 Mb/s baseband
networks
1.
2.
3.
4.
5.
6.
- "Cisco SFP Optics For Gigabit Ethernet Applications". Cisco Systems. Retrieved 2010-06-01.
7.
8.
9.
- IEEE. "Part 3: Carrier Sense Multiple Access with Collision Detection (CSMA/CD) access method and Physical Layer
specifications". SECTION TWO: This section includes Clause21 through Clause 33 and Annex 22A through Annex 33E.
Retrieved 2010-02-18.
Further reading
Norris, Mark, Gigabit Ethernet Technology and Applications, Artech House, 2002. ISBN 1-58053-505-4
External links
IEEE 802.3
IEEE and Gigabit Ethernet Alliance Announce Formal Ratification of gigabit Ethernet Over Copper Standard Announcement from IEEE June 28, 1999
In computer networking, 100 Gigabit Ethernet (or 100GbE) and 40 Gigabit Ethernet (or 40GbE) refers to
various technologies for transmitting Ethernet frames at a rates of 100 or 40 gigabits per second (100 to 40 Gbit/s), first
defined by the IEEE 802.3ba-2010 standard.[1]
Another variant, 802.3bg, was added in March 2011. There is an active task force 802.3bj [2] working on a four
lane backplane and copper 100 Gbit/s standard, and also the 802.3bm task force [3]working on a standard for lower cost
100 Gbit/s optical physical interfaces.
Contents
1 History
2 Standards
3.1 100GBASE-CR10
3.2 100GBASE-CR4
3.3 100GBASE-SR10
3.4 100GBASE-SR4
3.5 100GBASE-LR4
3.6 100GBASE-ER4
3.7 100GBASE-KR4
3.8 100GBASE-KP4
4.1 40GBASE-CR4
4.2 40GBASE-KR4
4.3 40GBASE-SR4
4.4 40GBASE-LR4
4.5 40GBASE-ER4
4.6 40GBASE-FR
4.7 40GBASE-T
5 Chip-to-Chip/Chip-to-Module Interfaces
5.1 CAUI
5.2 CAUI-4
6 Connectors
6.1 QSFP+
6.2 MPO
8 Products
8.1 Backplane
8.5 Compatibility
9.2 Products
10 See also
11 References
12 Further reading
13 External links
History
On July 18, 2006 a call for interest for a High Speed Study Group (HSSG) to investigate new standards for high
speed Ethernet was held at the IEEE 802.3 plenary meeting in San Diego. [4]
The first 802.3 HSSG study group meeting was held in September 2006. [5]
In June 2007, a trade group called "Road to 100G" was formed after the NXTcomm trade show in Chicago.[6]
On December 5, 2007 the Project Authorization Request (PAR) for the P802.3ba 40Gbit/s and 100Gbit/s
Ethernet Task Force was approved with the following project scope: [7]
The purpose of this project is to extend the 802.3 protocol to operating speeds of 40 Gb/s and 100 Gb/s in order
to provide a significant increase in bandwidth while maintaining maximum compatibility with the installed base of 802.3
interfaces, previous investment in research and development, and principles of network operation and management. The
project is to provide for the interconnection of equipment satisfying the distance requirements of the intended
applications.
The 802.3ba task force met for the first time in January 2008. [8] This standard was approved at the June 2010
IEEE Standards Board meeting under the name IEEE Std 802.3ba-2010. [9]
The first 40 Gbit/s Ethernet Single-mode Fibre PMD study group meeting was held in January 2010 and on
March 25, 2010 the P802.3bg Single-mode Fibre PMD Task Force was approved for the 40 Gbit/s serial SMF PMD.
The scope of this project is to add a single-mode fiber Physical Medium Dependent (PMD) option for serial 40
Gb/s operation by specifying additions to, and appropriate modifications of, IEEE Std 802.3-2008 as amended by the
IEEE P802.3ba project (and any other approved amendment or corrigendum).
On June 17, 2010 the IEEE 802.3ba standard was approved
[1][10]
Standards
The IEEE 802.3 working group is concerned with the maintenance and extension of the Ethernet data
communications standard. Additions to the 802.3 standard [13] are performed by task forces which are designated by one
or two letters. For example the 802.3z task force drafted the original gigabit Ethernet standard.
802.3ba is the designation given to the higher speed Ethernet task force which completed its work to modify the
802.3 standard to support speeds higher than 10 Gbit/s in 2010.
The speeds chosen by 802.3ba were 40 and 100 Gbit/s to support both end-point and link aggregation needs.
This was the first time two different Ethernet speeds were specified in a single standard. The decision to include both
speeds came from pressure to support the 40 Gbit/s rate for local server applications and the 100 Gbit/s rate for internet
backbones. The standard was announced in July 2007 [14] and was ratified on June 17, 2010.[9]
The 40/100 Gigabit Ethernet standards encompass a number of different Ethernet physical layer (PHY)
specifications. A networking device may support different PHY types by means of pluggable modules. Optical modules
are not standardized by any official standards body but are in multi-source agreements (MSAs). One agreement that
supports 40 and 100 Gigabit Ethernet is the C Form-factor Pluggable (CFP) MSA[15] which was adopted for distances of
100+ meters. QSFP and CXP connector modules support shorter distances. [16]
The standard supports only full-duplex operation.[17] Other electrical objectives include:
Preserve the 802.3 / Ethernet frame format utilizing the 802.3 MAC
Support a bit error ratio (BER) better than or equal to 1012 at the MAC/PLS service interface
Provide Physical Layer specifications (PHY) for operation over single-mode optical fiber (SMF), laser
optimized multi-mode optical fiber (MMF) OM3 and OM4, copper cable assembly, and backplane.
The following nomenclature was used for the physical layers: [18]
Physical layer
40 Gigabit
Ethernet
100 Gigabit
Ethernet
Backplane
40GBASE-KR4
Improved Backplane
100GBASE-KR4
40GBASE-CR4
40GBASE-T
100GBASE-KP4
40GBASE-SR4
40GBASE-FR
10 km over SMF
40GBASE-LR4
40 km over SMF
100GBASE-CR10
100GBASE-SR10
100GBASE-LR4
100GBASE-ER4
The 100 m laser optimized multi-mode fiber (OM3) objective was met by parallel ribbon cable with 850 nm
wavelength 10GBASE-SR like optics (40GBASE-SR4 and 100GBASE-SR10). The backplane objective with 4 lanes of
10GBASE-KR type PHYs (40GBASE-KR4). The copper cable objective is met with 4 or 10 differential lanes using SFF8642 and SFF-8436 connectors. The 10 and 40 km 100 Gbit/s objectives with four wavelengths (around 1310 nm) of
25 Gbit/s optics (100GBASE-LR4 and 100GBASE-ER4) and the 10 km 40 Gbit/s objective with four wavelengths (around
1310 nm) of 10 Gbit/s optics (40GBASE-LR4).[19]
In January 2010 another IEEE project authorization started a task force to define a 40 Gbit/s serial single-mode
optical fiber standard (40GBASE-FR). This was approved as standard 802.3bg in March 2011. [11] It used 1550 nm optics,
had a reach of 2 km and was capable of receiving 1550 nm and 1310 nm wavelengths of light. The capability to receive
1310 nm light allows it to inter-operate with a longer reach 1310 nm PHY should one ever be developed. 1550 nm was
chosen as the wavelength for 802.3bg transmission to make it compatible with existing test equipment and infrastructure.
[20]
In December 2010, a 10x10 Multi Source Agreement (10x10 MSA) began to define an optical Physical Medium
Dependent (PMD) sublayer and establish compatible sources of low-cost, low-power, pluggable optical transceivers
based on 10 optical lanes at 10 gigabits/second each. [21] The 10x10 MSA was intended as a lower cost alternative to
100GBASE-LR4 for applications which do not require a link length longer than 2 km. It was intended for use with
standard single mode G.652.C/D type low water peak cable with ten wavelengths ranging from 1523 to 1595 nm. The
founding members were Google, Brocade Communications, JDSU and Santur.[22] Other member companies of the 10x10
MSA included MRV, Enablence, Cyoptics, AFOP, OPLINK, Hitachi Cable America, AMS-IX, EXFO, Huawei, Kotura,
Facebook and Effdon when the 2 km specification was announced in March 2011. [23] The 10X10 MSA modules were
intended to be the same size as the C Form-factor Pluggable specifications.
There are currently two projects in 802.3 underway to specify additional PHYs. The 802.3bj task force is working
to produce 100 Gbit/s 4x25G PHYs for backplane and twin-ax cable (100GBASE-KR4, 100GBASE-KP4 and 100GBASECR4). The 802.3bm task force is working to produce lower cost optical PHYs. The detailed objectives for these projects
can be found on the 802.3 website.
100GBASE-CR10
100GBASE-CR10 ("copper") is a port type for twin-ax copper cable. Its Physical Coding Sublayer 64b/66b PCS
is defined in IEEE 802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 85. It uses ten lanes of twin-ax
cable delivering serialized data at a rate of 10.3125 Gbit/s per lane.[13]
100GBASE-CR4
100GBASE-CR4 ("copper") is a port type for twin-ax copper cable. Its Physical Coding Sublayer 64b/66b PCS is
defined in IEEE 802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 92 of 802.3bj. It uses four lanes of
twin-ax cable delivering serialized data at a rate of 25.78125 Gbit/s per lane.[2]
100GBASE-SR10
100GBASE-SR10 ("short range") is a port type for multi-mode fiber and uses 850 nm lasers. Its Physical Coding
Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 86. It
uses ten lanes of multi-mode fiber delivering serialized data at a rate of 10.3125 Gbit/s per lane.[13]
100GBASE-SR4
100GBASE-SR4 ("short range") is a port type for multi-mode fiber being defined in P802.3bm and uses 850 nm
lasers. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and its Physical Medium
Dependent PMD in Clause 95. It uses four lanes of multi-mode fiber delivering serialized RS-FEC encoded data at a rate
of 25.78125 Gbit/s per lane.[3]
100GBASE-LR4
100GBASE-LR4 ("long range") is a port type for single-mode fiber and uses four lasers using four wavelengths
around 1300 nm. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and its Physical Medium
Dependent PMD in Clause 88. Each wavelength carries data at a rate of 25.78125 Gbit/s.[13]
100GBASE-ER4
100GBASE-ER4 ("extended range") is a port type for single-mode fiber and uses four lasers using four
wavelengths around 1300 nm. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and
its Physical Medium Dependent PMD in Clause 88. Each wavelength carries data at a rate of 25.78125 Gbit/s.[13]
100GBASE-KR4
100GBASE-KR4 is a port type for backplanes. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE
802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 93 of 802.3bj. It delivers Reed Solomon encoded
serialized data at a rate of 25.78125 Gbit/s per lane over four lanes of up to one meter of backplane. The Reed Solomon
forward error correction is defined in Clause 91. [2]
100GBASE-KP4
100GBASE-KP4 is a port type for backplanes. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE
802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 94. The data is further encoded by the Reed
Solomon forward error correction is defined in Clause 91 and the four level amplitude modulation is defined in Clause 94
of 802.3bj. 100GBASE-KP4 uses more power than 100GBASE-KR4 but is designed to work on lower cost and legacy
backplanes.[2]
40GBASE-KR4
40GBASE-KR4 is a port type for backplanes. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE
802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 84. It uses four lanes of backplane delivering
serialized data at a rate of 10.3125 Gbit/s per lane.[13]
40GBASE-SR4
40GBASE-SR4 ("short range") is a port type for multi-mode fiber and uses 850 nm lasers. Its Physical Coding
Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 86. It
uses four lanes of multi-mode fiber delivering serialized data at a rate of 10.3125 Gbit/s per lane. 40GBASE-SR4 has a
reach of 100m on OM3 and 150m on OM4. There is a longer range variant 40GBASE-eSR4 with a reach of 300m on
OM3 and 400m on OM4. This extended reach is equivalent to the reach of 10GBASE-SR. [24]
40GBASE-LR4
40GBASE-LR4 ("long range") is a port type for single-mode fiber and uses 1300 nm lasers. Its Physical Coding
Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 87. It
uses four wavelengths delivering serialized data at a rate of 10.3125 Gbit/s per wavelength.[13]
40GBASE-ER4
40GBASE-ER4 ("extended range") is a port type for single-mode fiber being defined in P802.3bm and uses
1300 nm lasers. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE 802.3 Clause 82 and its Physical Medium
Dependent PMD in Clause 87. It uses four wavelengths delivering serialized data at a rate of 10.3125 Gbit/s per
wavelength.[3]
40GBASE-FR
40GBASE-FR is a port type for single-mode fiber. Its Physical Coding Sublayer 64b/66b PCS is defined in IEEE
802.3 Clause 82 and its Physical Medium Dependent PMD in Clause 89. It uses 1550 nm optics, has a reach of 2 km
and is capable of receiving 1550 nm and 1310 nm wavelengths of light. The capability to receive 1310 nm light allows it
to inter-operate with a longer reach 1310 nm PHY should one ever be developed. 1550 nm was chosen as the
wavelength transmission to make it compatible with existing test equipment and infrastructure. [13]
40GBASE-T
40GBASE-T is a port type for 4-pair balanced twisted-pair Cat.8 copper cabling being defined in P802.3bq. [25]
Chip-to-Chip/Chip-to-Module Interfaces
CAUI
CAUI is a 100 Gbit/s 10 lane electrical interface defined in 802.3ba. [1]
CAUI-4
CAUI-4 is a 100 Gbit/s 4 lane electrical interface being defined in 802.3bm. [3]
Connectors
QSFP+
The QSFP+ connector is specified for use with the 40GBASE-CR4 PHY, see Figure 8520 in the 802.3 spec. [1]
MPO
The 40GBASE-SR4 and 100GBASE-SR10 PHYs use the Multiple-Fiber Push-On/Pull-off (MPO) connector, see
subclause 86.10.3.3 of the 802.3 spec.[1]
Cisco has the CPAK optical module that uses the four lane CEI-28G-VSR electrical interface. The QSFP28
module also uses this electrical interface.
[27][28]
Products
Backplane
NetLogic Microsystems announced backplane modules in October 2010.[30]
Copper cables
Quellan announced a test board in 2009. [31]
Multimode fiber
In 2009, Mellanox[32] and Reflex Photonics[33] announced modules based on the CFP agreement.
Compatibility
Optical fiber IEEE 802.3ba implementations were not compatible with the numerous 40 Gbit/s and 100 Gbit/s line
rate transport systems because they had different optical layer and modulation formats. In particular, existing 40 Gbit/s
transport solutions that used dense wavelength-division multiplexing to pack four 10 Gbit/s signals into one optical
medium were not compatible with the IEEE 802.3ba standard, which used either coarse WDM in 1310 nm wavelength
region with four 25 Gbit/s or four 10 Gbit/s channels, or parallel optics with four or ten optical fibers per direction. [citation
needed]
Xena Networks demonstrated test equipment at the Technical University of Denmark in January 2011.[45][46]
to reduce the number of optical wavelengths ("lambdas") used and the need to light new fiber
to provide cheaper wholesale, internet peering and data center interconnect connectivity
to skip the relatively expensive 40 Gbit/s technology and move directly from 10 Gbit/s to 100 Gbit/s
Considering that 100GbE technology is natively compatible with Optical Transport Network (OTN) hierarchy and
there is no separate adaptation for SONET/SDH and Ethernet networks, it was widely believed [by
whom?]
that 100GbE
technology adoption will be driven by products in all network layers, from transport systems to edge routers and
datacenter switches. Nevertheless, in 2011 components for 100GE networks were expensive and most vendors entering
this market relied on internal R&D projects and extensive cooperation with other companies. [citation needed]
Products
Designing routers or switches supporting 100 Gbit/s interfaces is difficult. One reason is the need to process a
100 Gbit/s stream of packets at line rate without reordering within IP/MPLS microflows. As of 2011, most components in
the 100 Gbit/s packet processing path (PHY chips, NPUs, memories) were not readily available off-the-shelf or require
extensive qualification and co-design. Another problem is related to the low-output production of 100 Gbit/s optical
components, which were also not easily available especially in pluggable, long-reach or tunable laser flavors.
Alcatel-Lucent
In November 2007 Alcatel-Lucent held the first field trial of 100 Gbit/s optical transmission. Completed over a
live, in-service 504-km portion of the Verizon network, it connected the Florida cities of Tampa and Miami. [49] 100GbE
interfaces for the 7450 ESS/7750 SR service routing platform were first announced in June 2009, with field trials with
Verizon,[50] T-Systems and Portugal Telecom following in JuneSeptember 2010. In September 2009 Alcatel-Lucent
combined the 100G capabilities of its IP routing and optical transport portfolio in an integrated solution called Converged
Backbone Transformation.[51]
In June 2011, Alcatel-Lucent announced a packet processing architecture called FP3, advertised for 400 Gbit/s
rates.
[52]
Arista
Arista Networks announced its 7500E switch with up to 96 100GbE ports in April 2013. [55]
Brocade
In September 2010, Brocade Communications Systems announced their first 100GbE products based on the
former Foundry Networks hardware (MLXe). [56] In June 2011, the new product went live at AMS-IX traffic exchange point
in Amsterdam.[57]
Cisco
Cisco Systems and Comcast announced their 100GbE trials in June 2008, [58] however it is doubtful this
transmission could approach 100 Gbit/s speeds when using a 40 Gbit/s per slot CRS-1 platform for packet processing.
Cisco's first deployment of 100GbE at AT&T and Comcast occurred in April 2011. [59] Later in the same year, Cisco tested
the 100GbE interface between CRS-3 and a new generation of their ASR9K edge router.[60]
Extreme Networks
Extreme Networks announced its first 100GbE product on November 13, 2012, a four-port 100GbE module for
the BlackDiamond X8 core switch.[61]
Huawei
In October 2008, the Chinese vendor Huawei presented their first 100GbE interface for their NE5000e router.
[62]
In September 2009, Huawei also demonstrated an end-to-end 100&Gbit/s link. [63] It was mentioned that Huawei's
products had the self-developed NPU "Solar 2.0 PFE2A" onboard and was using pluggable optics in CFP form-factor. In
a mid-2010 product brief, the NE5000e linecards were given commercial name LPUF-100 and credited with using two
Solar-2.0 NPUs per 100GbE port in opposite (ingress/egress) configuration. [64] Nevertheless, in October 2010, the
company referenced shipments of NE5000e to Russian cell operator "Megafon" as "40Gbps/slot" solution, with
"scalability up to" 100Gbit/s.[65]
In April 2011, Huawei announced that the NE5000e was updated to carry 2x100GbE interfaces per slot using
LPU-200 linecards.[66] In a related solution brief, Huawei reported 120 thousand Solar 1.0 integrated circuits shipped to
customers, but no Solar 2.0 numbers were given. [67] Following the August 2011 trial in Russia, Huawei reported paying
100 Gbit/ DWDM customers, but no 100GbE shipments on NE5000e.[68]
Juniper
Juniper Networks announced 100GbE for its T-series routers in June 2009. [69] The 1x100GbE option followed in
Nov 2010, when a joint press release with academic backbone network Internet2marked the first production 100GbE
interfaces going live in real network.[70] Later in the same year, Juniper demonstrated 100GbE operation between core (Tseries) and edge (MX 3D) routers. [71]Juniper, in March 2011, announced first shipments of 100GbE interfaces to a major
North American service provider (Verizon [72]). In April 2011, Juniper deployed a 100GbE system to the UK network
operator JANET.[73] In July 2011, Juniper announced 100GbE with Australian ISP iiNet on their T1600 routing platform. [74]
In March 2012, Juniper Networks started shipping the MPC3E line card for the MX router, a 10GbE CFP MIC,
and a 100GbE LR4 CFP optics. In Spring 2013, Juniper Networks announced the availability of the MPC4E line card for
the MX router that includes 2 100GbE CFP slots and 8 10GbE SFP+ interfaces.
Dell
Dell's Force10 switches support 40 Gbit/s interfaces. These 40 Gbit/s fiber-optical interfaces using QSFP+
transceivers can be found on the Z9000 distributed core switches, S4810 and S4820 [75]as well as the blade-switches MXL
and the IO-Aggregator. The Dell PowerConnect 8100 series switches also offer 40 Gbit/s QSFP+ interfaces.[76]
Chelsio
In June 2013, Chelsio Communications, announced 40 Gbit/s Ethernet network adapters based on the fifth
generation of its Terminator architecture.[77]
See also
Energy Efficient Ethernet
Ethernet Alliance
Interconnect bottleneck
Optical communication
Optical interconnect
References[edit]
^ :a b c d e "IEEE P802.3ba 40Gb/s and 100Gb/s Ethernet Task Force". official web site. IEEE. June 19, 2010.
1.
^ Jump
up
to:a b c d e "100
Gb/s
Backplane
and
Copper
Cable
Task
Force". official
web
site.
^ Jump up to:a b c d e "40 Gb/s and 100 Gb/s Fiber Optic Task Force".official web site. IEEE.
4.
- "IEEE Forms Higher Speed Study Group to Explore the Next Generation of Ethernet Technology". 2006-0725.
5.
- "IEEE 802.3 Higher Speed Study Group". IEEE802.org. Retrieved December 17, 2011.
6.
- Jeff Caruso (June 21, 2007). "Group pushes 100 Gigabit Ethernet: The 'Road to 100G' Alliance is
born". Network World. Retrieved June 6, 2011.
7.
- "Project Authorization Request Approval notification: Approcal of P802.3ba". IEEE Standards Association
Standards Board. December 5, 2007. Retrieved June 6, 2011.
8.
- Caruso, Jeff (2008-01-15). "Standardization work on next Ethernet gets under way". NetworkWorld.
9.
^ Jump up to:a b "IEEE P802.3ba 40Gb/s and 100Gb/s Ethernet Task Force". 2010-06-21.
10.
- "IEEE 802.3ba standard released". Help Net Security web site. June 21, 2010. Retrieved June 24, 2011.
"The IEEE 802.3ba standard, ratified June 17, 2010, ..."
11.
^ Jump up to:a b "IEEE P802.3bg 40Gb/s Ethernet: Single-mode Fibre PMD Task Force". official task force
web site. IEEE 802. April 12, 2011. Retrieved June 7, 2011.
12.
- "P802.3bq PAR".
13.
14.
- Reimer, Jeremy (2007-07-24). "New Ethernet standard: not 40Gbps, not 100, but both". ars technica.
15.
- "CFP Multi-Source Agreement". official web site.Archived from the original on September 27, 2009.
Retrieved June 24, 2011.
16.
Greg Hankins (October 20, 2009). "IEEE P802.3ba 40 GbE and 100 GbE Standards Update" (PDF). North
American Network Operators' Group (NANOG) 47 Presentations. Retrieved June 24, 2011.
17.
- John D'Ambrosia. "IEEE P802.3ba Objectives".Archived from the original on September 27, 2009. Retrieved
September 25, 2009.
18.
- Ilango Ganga (May 13, 2009). "Chief Editor's Report".IEEE P802.3ba 40Gb/s and 100Gb/s Ethernet Task
Force public record. p. 8. Retrieved June 7, 2011.
19.
- Ilango Ganga; Brad Booth; Howard Frazier; Shimon Muller; Gary Nicholl (May 13, 2008). "IEEE P802.3ba
40Gb/s and 100Gb/s Ethernet Task Force, May 2008 Meeting".
20.
21.
- "10 x 10 MSA Low Cost 100 GB/s Pluggable Optical Transceiver". official web site. 10x10 multi-source
agreement. Retrieved June 24, 2011.
22.
- "Leading Industry Peers Join Forces to Develop Low-Cost 100G Multi-Source Agreement". Businesswire
news release. December 7, 2010. Retrieved June 24, 2011.
23.
- "10X10 MSA Ratifies Specification for Low Cost 100 Gb/s 2 Kilometer Links". News release (10x10 MSA).
March 4, 2011. Retrieved June 24, 2011.
24.
25.
26.
"CFP MSA".
27.
28.
- "Multi-Vendor Interoperability Testing of CFP2, CPAK and QSFP28 with CEI-28G-VSR and CEI-25G-LR
Interface During ECOC 2013 Exhibition".
29.
- Daniel Dove. "4X25G Optical Modules and Future Optics". Archived from the original on 2013-07-11.
Retrieved 2013-07-04.
30.
- "NetLogic Microsystems Announces Industry's First Dual-Mode Quad-Port 10GBASE-KR and 40GBASEKR4 Backplane PHY with Energy Efficient Ethernet". News release (NetLogic Microsystems). October 13, 2010.
Retrieved June 24, 2013.
31.
- "Quellan QLx411GRx 40G Evaluation Board". Archivedfrom the original on September 27, 2009. Retrieved
September 25, 2009.
32.
- "Mellanox Technologies". Archived from the original on September 27, 2009. Retrieved September 25, 2009.
33.
- "InterBOARD CFP 100GBASE-SR10 Parallel Optical Module". commercial web site. Reflex Photonics Inc.
Archived from the original on September 27, 2009. Retrieved June 7, 2011.
34.
- "Finisar Corporation Finisar First to Demonstrate 40 Gigabit Ethernet LR4 CFP Transceiver Over 10 km of
Optical Fiber at ECOC". Archived from the original on September 27, 2009. Retrieved September 25, 2009.
35.
36.
- "Hitachi and Opnext unveil receiver for 100GbE and demo 10 km transmission over SMF". Retrieved
October 26, 2009.
37.
38.
- "Avago Technologies, Infinera & Ixia to demo the first 100 Gig Ethernet". Archived from the original on 201203-09. Retrieved 7 March 2012.
39.
- "Ixia First to Offer 100 GE Testing Capability". News release (Ixia). September 29, 2008. Retrieved June 7,
2011.
40.
- "40 Gb/s and 100 Gb/s Testing: Overview". commercial web site. Ixia. Retrieved June 7, 2011.
41.
- "Discovery Semiconductors 100 Gb Ethernet (4 x 25 Gb/s) Quad PIN-TIA Optical Receiver". commercial
web site. Retrieved June 7, 2011.
42.
- "JDSU Introduces Industrys Most Robust 100 Gigabit Ethernet Test Suite". News release. JDS Uniphase.
August 19, 2009. Retrieved June 7, 2011.
43.
- "40/100 GbE: Testing the next generation of high speed Ethernet". commercial web site. Spirent
Communications. Retrieved June 7, 2011.
44.
- "EXFO and Opnext Achieve Full Interoperability, Successfully Testing IEEE-Compliant 100 Gigabit Ethernet
Optics". News release. January 11, 2010. Retrieved June 7, 2011.
45.
- "Workshop on 100 Gigabit Ethernet a huge success".DTU news (Technical University of Denmark). February
2, 2011. Retrieved June 7, 2011.
46.
- Torben R. Simonsen (January 26, 2011). "Dansk virksomhed klar med test til 100 Gb ethernet". Elektronik
Branchen. Retrieved June 7, 2011. (Danish)
47.
48.
49.
- "Verizon Successfully Completes Industrys First Field Trial of 100 Gbps Optical Network Transmission".
50.
51.
52.
- "Alcatel-Lucent's FP3 network processor routes at 400Gbps". Press release. June 29, 2011. Retrieved June
24, 2013.
53.
- David Goldman (May 21, 2012). "How Alcatel-Lucent made the Internet 5 times faster". CNN Money.
Retrieved June 24, 2013.
54.
- "100 Gigabit Ethernet (100GE): Services unleashed at speed". Company web site. Archived from the original
on November 16, 2012. Retrieved June 24, 2013.
55.
- Jim Duffy (May 1, 2013). "Arista heading off Cisco/Insieme at 100G SDNs?". Network World. Retrieved May
24, 2013.
56.
57.
58.
59.
- Matsumoto, Craig (April 11, 2011). "AT&T, Comcast Go Live With 100G". Light Reading. Retrieved
December 17, 2011.
60.
- Liu, Stephen (July 25, 2011). "Cisco Live! Showing Off 100GbE on CRS-3 and ASR 9000 Series".
blogs.cisco.com. Retrieved December 17, 2011.
61.
- Duffy, Jim (November 13, 2012). "Extreme joins Cisco, Brocade, Huawei at 100G". Network World. p. 1.
Retrieved January 18, 2013.
62.
63.
64.
65.
66.
- "Huawei Unveils the World's First 200G High-Speed Line Card for Routers".
67.
68.
69.
- "Juniper networks introduces breakthrough 100 gigabit ethernet interface for t series routers".
70.
71.
- "Juniper Demonstrates Industry's First Live 100G Traffic From the Network Core to Edge".
72.
73.
74.
75.
76.
77.
- "Chelsio Delivers 40Gb Ethernet Adapter (40GbE), Sets new performance bar for high speed
Ethernet". Press release. June 11, 2013. Archived from the original on 2013-06-22. Retrieved June 20, 2013.
Terabit
possible
speed
of
Ethernet
above 100
Gigabit
[1]
Ethernet. Facebook and Google, among other companies, have expressed a need for TbE. Some think that400 Gigabit
Ethernet is a more practical goal. [2] In 2011 researchers predicted Terabit Ethernet in 2015, and 100 Terabit Ethernet by
2020.[3] UCSB attracted help from Agilent Technologies,Google, Intel, Rockwell Collins, and Verizon Communications to
help with the research.[4]
The IEEE announced the formation of "IEEE 802.3 Industry Connections Ethernet Bandwidth Assessment Ad
Hoc," to investigate the business needs for short and long term bandwidth requirements. They planned to formally
announce reports of findings the first half of 2012. [5] [6]
IEEE 802.3's "400 Gb/s Ethernet Study Group" started working on the 400 Gbit/s generation standard in March
[7]
1 See also
2 References
3 Further reading
4 External
See also[edit]
Ethernet Alliance
Optical interconnect
Interconnect bottleneck
Optical communication
References[edit]
1.
- Feldman, Michael (Feb 3, 2010). "Facebook Dreams of Terabit Ethernet". HPCwire. Tabor Communications,
Inc.
2.
- Matsumoto, Craig (March 5, 2010). "Dare We Aim for Terabit Ethernet?". Light Reading. UBM TechWeb,.
3.
- "UCSBs Professor Daniel Blumenthal to Address the Road to Terabit Ethernet at the Ethernet Technology
Summit". University of California, Santa Barbara. February 18, 2011. Retrieved May 23, 2013.
4.
- Craig Matsumoto (October 26, 2010). "The Terabit Ethernet Chase Begins". Light Reading. Retrieved 15
Dec 2011.
5.
- Stephen Lawson (May 9, 2011). "IEEE Seeks Data on Ethernet Bandwidth Needs". PC World. Retrieved
May 23, 2013.
6.
- Max Burkhalter Brafton (12 May 2011). "Terabit Ethernet could be on its way". Perle. Retrieved 15 Dec 2011.
7.
- "400 Gb/s Ethernet Study Group". Group web site. IEEE 802.3. Retrieved May 23, 2013.
8.
- Jim Duffy (April 2, 2013). ""Tsunami" of bandwidth demand pushes IEEE 400G Ethernet standards process".
NetworkWorld. Retrieved May 23, 2013.
Further reading[edit]
Chris Jablonski. "Researchers to develop 1 Terabit Ethernet by 2015". ZD Net. Retrieved 9 Oct 2011.
Iljitsch van Beijnum (Aug 2011). "Speed matters: how Ethernet went from 3Mbps to 100Gbps... and beyond". Ars
Technica. Retrieved 9 Oct 2011.
Rick Merritt (9 May 2011). "IEEE Looks beyond 100G Ethernet". The Cutting Edge. Retrieved 9 Oct 2011.
Stephen Lawson (2 Feb 2010). "Facebook Sees Need for Terabit Ethernet". PC World. Retrieved 15 Dec 2011.
IEEE Reports
"DQPSK for Terabit Ethernet in the 1310 nm band". July 2011. ISBN 978-1-4244-5730-4.
External[edit]
Mellor, Chris (Feb 15, 2009). "Terabit Ethernet possibilities". The Register.
Duffy, Jim (Apr 20, 2009). "100 Gigabit Ethernet: Bridge to Terabit Ethernet". Network World.
Fleishman, Glenn (Feb 13, 2009). "Terabit Ethernet becomes a photonic possibility". Ars Technica. Cond Nast.