100% found this document useful (1 vote)
419 views5 pages

ARCNET

ARCNET was developed in 1976 as a local area network protocol for microcomputers and became popular in the 1980s for office automation. It used a token-passing protocol over coaxial cable in a star topology and allowed up to 255 nodes per network. While ARCNET was competitive with early Ethernet networks, Ethernet eventually dominated as it offered higher speeds, easier cabling, and broader standards support.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
419 views5 pages

ARCNET

ARCNET was developed in 1976 as a local area network protocol for microcomputers and became popular in the 1980s for office automation. It used a token-passing protocol over coaxial cable in a star topology and allowed up to 255 nodes per network. While ARCNET was competitive with early Ethernet networks, Ethernet eventually dominated as it offered higher speeds, easier cabling, and broader standards support.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

3/26/2016

ARCNETWikipedia,thefreeencyclopedia

ARCNET
FromWikipedia,thefreeencyclopedia

AttachedResourceComputerNETwork(abbreviatedARCNETorARCnet)isacommunications
protocolforlocalareanetworks.ARCNETwasthefirstwidelyavailablenetworkingsystemfor
microcomputersitbecamepopularinthe1980sforofficeautomationtasks.Itwaslaterappliedto
embeddedsystemswherecertainfeaturesoftheprotocolareespeciallyuseful.

Contents
1
2
3
4
5

History
Description
Seealso
References
Externallinks

History
ARCNETwasdevelopedbyprincipaldevelopmentengineerJohnMurphyatDatapointCorporationin
1976underVictorPoor,andannouncedin1977.[1]Itwasoriginallydevelopedtoconnectgroupsoftheir
Datapoint2200terminalstotalktoashared8"floppydisksystem.Asmicrocomputerstookoverfromthe
Datapoint,ARCNETwasrepurposedasLAN.ItwasthefirstlooselycoupledLANbasedclustering
solution,makingnoassumptionsaboutthetypeofcomputersthatwouldbeconnected.Thiswasincontrast
tocontemporarylargerandmoreexpensivecomputersystemssuchasDECnetorSNA,wherea
homogeneousgroupofsimilarorproprietarycomputerswereconnectedasacluster.
ThetokenpassingbusprotocolofthatI/Odevicesharingnetworkwassubsequentlyappliedtoallowing
processingnodestocommunicatewitheachotherforfileservingandcomputingscalabilitypurposes.An
applicationcouldbedevelopedinDATABUS,Datapoint'sproprietaryCOBOLlikelanguageanddeployed
onasinglecomputerwithdumbterminals.Whenthenumberofusersoutgrewthecapacityoftheoriginal
computer,additional'compute'resourcecomputerscouldbeattachedviaARCNET,runningthesame
applicationsandaccessingthesamedata.Ifmorestoragewasneeded,additionaldiskresourcecomputers
couldalsobeattached.Thisincrementalapproachbrokenewgroundandbytheendofthe1970s(before
thefirstcassettebasedIBMPCwasannouncedin1981)overtenthousandARCnetLANinstallationswere
incommercialusearoundtheworld,andDatapointhadbecomeaFortune500company.As
microcomputerstookovertheindustry,wellprovenandreliableARCNETwasalsoofferedasan
inexpensiveLANforthesemachines.
ARCNETremainedproprietaryuntiltheearlytomid1980s.Thisdidnotcauseconcernatthetime,as
mostnetworkarchitectureswereproprietary.Themovetononproprietary,opensystemsbeganasa
responsetothedominanceofInternationalBusinessMachines(IBM)anditsSystemsNetworkArchitecture
(SNA).In1979,theOpenSystemsInterconnectionReferenceModel(OSImodel)waspublished.Then,in
1980,Digital,IntelandXerox(theDIXconsortium)publishedanopenstandardforEthernetthatwassoon
adoptedasthebasisofstandardizationbytheIEEEandtheISO.IBMrespondedbyproposingTokenRing
https://en.wikipedia.org/wiki/ARCNET

1/5

3/26/2016

ARCNETWikipedia,thefreeencyclopedia

asanalternativetoEthernetbutkeptsuchtightcontroloverstandardizationthatcompetitorswerewaryof
usingit.ARCNETwaslessexpensivethaneither,morereliable,moreflexible,andbythelate1980sithad
amarketshareaboutequaltothatofEthernet.Tandy/RadioShackofferedARCNETasanapplicationand
filesharingmediumfortheirTRS80ModelII,Model12,Model16,Tandy6000,Tandy2000,Tandy1000
andTandy1200computermodels.[2][3]
WhenEthernetmovedfromcoaxialcabletotwistedpairandan"interconnectedstars"cablingtopology
basedonactivehubs,itbecamemuchmoreattractive.Easiercabling,combinedwiththegreaterrawspeed
ofEthernet(10Mbit/s,ascomparedwith2.5Mbit/sforARCnet)helpedtoincreaseEthernetdemand,and
asmorecompaniesenteredthemarketthepriceofEthernetstartedtofallandARCNET(andTokenRing)
volumestaperedoff.
Inresponsetogreaterbandwidthneeds,andthechallengeofEthernet,anewstandardcalledARCnetPlus
wasdevelopedbyDatapoint,andintroducedin1992.ARCnetPlusranat20Mbit/s,andwasbackward
compatiblewithoriginalARCnetequipment.However,bythetimeARCnetPlusproductswerereadyfor
themarket,Ethernethadcapturedthemajorityofthenetworkmarket,andtherewaslittleincentivefor
userstomovebacktoARCnet.Asaresult,veryfewARCnetPlusproductswereeverproduced.Thosethat
werebuilt,mainlybyDatapoint,wereexpensive,andhardtofind.
ARCNETwaseventuallystandardizedasANSIARCNET878.1.Itappearsthiswaswhenthename
changedfromARCnettoARCNET.Othercompaniesenteredthemarket,notablyStandardMicrosystems
whoproducedsystemsbasedonasingleVLSIchip,originallydevelopedascustomLSIforDatapoint,but
latermadeavailablebyStandardMicrosystemstoothercustomers.Datapointeventuallyfounditselfin
financialtroubleandeventuallymovedintovideoconferencingand(later)customprogramminginthe
embeddedmarket.
EventhoughARCNETisnowrarelyusedfornewgeneralnetworks,thediminishinginstalledbasestill
requiressupportanditretainsanicheinindustrialcontrol.[4]

Description
OriginalARCNETusedRG62/Ucoaxialcableof93impedanceandeitherpassiveoractivehubsina
starwiredbustopology.Atthetimeofitsgreatestpopularity,thiswasasignificantadvantageofARCNET
overEthernet.Astarwiredbuswasmucheasiertobuildandexpand(andwasmorereadilymaintainable)
thantheclumsylinearbusEthernetofthetime.The"interconnectedstars"cablingtopologymadeiteasyto
addandremovenodeswithouttakingdownthewholenetwork,andmucheasiertodiagnoseandisolate
failureswithinacomplexLAN.
AnothersignificantadvantageARCNEThadoverEthernetwascabledistance.ARCNETcoaxcableruns
couldextend610m(2,000ft)betweenactivehubsorbetweenanactivehubandanendnode,whilethe
RG58(50)'thin'Ethernetmostwidelyusedatthattimewaslimitedtoamaximumrunof180m(600ft)
fromendtoend.
ARCNEThadthedisadvantageofrequiringeitheranactiveorpassivehubbetweennodesiftherewere
morethantwonodesinthenetwork,whilethinEthernetallowednodestobespacedanywherealongthe
linearcoaxcable.However,ARCNETpassivehubswereveryinexpensive,beingcomposedofasimple,
small,unpoweredboxwithfourports,wiredtogetherwithnothingmorethanfourdiscreteresistors,sothe
disadvantagewasnotsignificant.
https://en.wikipedia.org/wiki/ARCNET

2/5

3/26/2016

ARCNETWikipedia,thefreeencyclopedia

Passivehubslimitedthedistancebetweenanodeandanactivehubto30m(100ft).Apassivehubcould
notbeconnecteddirectlytoanotherpassivehub.Unusedportsonbothtypesofhubshadtobeterminated
withaspecialconnector.
Toreducecosts,whilestillallowingcoverageoverawidearea,acommonpracticewastouseoneormore
interconnectedactivehubs,eachofwhichprovidedcoveragefornodesnomorethan60m(200ft)away.
Cablewasrunfromeachportoftheactivehubstoadifferentlocationnomorethan30m(100ft)away.A
passivehubwouldthenbeattachedtotheendofthecable,andcableswouldberunlocallyfromthe
passivehub,allowingconnectionofuptothreenodes.Inthisway,asingle8portactivehubcouldbeused
toconnect24networkeddevicesoveranareanotexceeding120m(400ft)indiameter.
ARCNETallowedonly255nodespernetwork.NodeIDsforLANworkstationsweretypicallysetbyDIP
switchesonthenetworkinterfacecard.Largernetworkswouldhavetobesplitintosmallernetworks,and
bridged.ThesmallnumberofpossiblenodesandtheneedtomanuallyconfigureIDswasadisadvantage
comparedwithEthernet,particularlyaslargeenterprisenetworksbecamecommon.
Tomediateaccesstothebus,ARCNET,likeTokenRing,usesatokenpassingscheme,ratherthanthe
carriersensemultipleaccessapproachofEthernet.Whenpeersareinactive,asingle"token"messageis
passedaroundthenetworkfrommachinetomachine,andnopeerisallowedtousethebusunlessithasthe
token.Ifaparticularpeerwishestosendamessage,itwaitstoreceivethetoken,sendsitsmessage,and
thenpassesthetokenontothenextstation.BecauseARCNETisimplementedasadistributedstar,the
tokencannotbepassedmachinetomachinearoundaring.Instead,eachnodeisassignedan8bitaddress
(usuallyviaDIPswitches),andwhenanewnodejoinsthenetworka"reconfig"occurs,whereineachnode
learnstheaddressofthenodeimmediatelyaboveit.Thetokenisthenpasseddirectlyfromonenodetothe
next.
Historically,eachapproachhaditsadvantages:ARCNETaddedasmalldelayonaninactivenetworkasa
sendingstationwaitedtoreceivethetoken,butEthernet'sperformancedegradeddrasticallyiftoomany
peersattemptedtobroadcastatthesametime,duetothetimerequiredfortheslowerprocessorsoftheday
toprocessandrecoverfromcollisions.ARCNEThadslightlylowerbestcaseperformance(viewedbya
singlestream),butwasmuchmorepredictable.ARCNETalsohastheadvantagethatitachieveditsbest
aggregateperformanceunderthehighestloading,approachingasymptoticallyitsmaximumthroughput.
WhilethebestcaseperformancewaslessthanEthernet,thegeneralcasewasequivalentandtheworstcase
wasdramaticallybetter.AnEthernetnetworkcouldcollapsewhentoobusyduetoexcessivecollisions.An
ARCNETwouldkeepongoingatnormal(orevenbetter)throughput.Throughputonamultinode
collisionbasedEthernetwaslimitedtobetween40%and60%ofbandwidthusage(dependingonsource).
Although2.5Mbit/sARCNETcouldatonetimeoutperforma10Mbit/sEthernetinabusyofficeonslow
processors,ARCNETultimatelygavewaytoEthernetasimprovedprocessorspeedsreducedtheimpactof
collisionsonoverallthroughput,andEthernetcostsdropped.
Intheearly1980sARCNETwasmuchcheaperthanEthernet,inparticularforPCs.Forexample,in1985
SMCsoldARCNETcardsforaroundUS$300whilstanUngermannBassEthernetcardplustransceiver
couldcostUS$500.
AnothersignificantdifferenceisthatARCNETprovidesthesenderwithaconcreteacknowledgment(or
not)ofsuccessfuldeliveryatthereceivingendbeforethetokenpassesontothenextnode,permittingmuch
fasterfaultrecoverywithinthehigherlevelprotocols(ratherthanhavingtowaitforatimeoutonthe

https://en.wikipedia.org/wiki/ARCNET

3/5

3/26/2016

ARCNETWikipedia,thefreeencyclopedia

expectedreplies).ARCnetalsodoesn'twastenetworktimetransmittingtoanodenotreadytoreceivethe
message,sinceaninitialinquiry(doneathardwarelevel)establishesthattherecipientisableandreadyto
receivethelargermessagebeforeitissentacrossthebus.
OnefurtheradvantagethatARCNETenjoyedovercollisionbasedEthernetisthatitguaranteesequitable
accesstothebusbyeveryoneonthenetwork.Althoughitmighttakeashorttimetogetthetoken
dependingonthenumberofnodesandthesizeofthemessagescurrentlybeingsentabout,youwillalways
receiveitwithinapredictablemaximumtimethusitisdeterministic.ThismadeARCNETanidealreal
timenetworkingsystem,whichexplainsitsuseintheembeddedsystemsandprocesscontrolmarkets.
TokenRinghassimilarqualities,butismuchmoreexpensivetoimplementthanARCNET.
InspiteofARCNET'sdeterministicoperationandhistoricsuitabilityforrealtimeenvironmentssuchas
processcontrol,thegeneralavailabilityofswitchedgigabitEthernetandQualityofservicecapabilitiesin
EthernetswitcheshasallbuteliminatedARCNETtoday.
AtfirstthesystemwasdeployedusingRG62/Ucoaxialcable(commonlyusedinIBMmainframe
environmentstoconnect3270terminalsandcontrollers),butlateraddedsupportfortwistedpairandfibre
media.AtARCNET'slowerspeeds(2.5Mbit/s),Cat3cableisgoodenoughtorunARCNET.Some
ARCNETtwistedpairproductssupportedcablerunsover2000'onstandardCat3cable,farbeyond
anythingEthernetcoulddoonanykindofcoppercable.
Intheearly1990s,ThomasConradCorporationdevelopeda100Mbit/stopologycalledTCNSbasedon
theARCNETprotocol,whichalsosupportedRG62,twistedpair,andfiberopticmedia.[5]TCNSenjoyed
somesuccessuntiltheavailabilityoflowercost100Mbit/sEthernetputanendtothegeneraldeployment
ofARCNETasaLANprotocol.
However,becauseofitssimple,robustnature,ARCNETcontrollersarestillsoldandusedinindustrial,
embedded,andautomotiveapplications.

Seealso
Listofdevicebandwidths
RFC1201TransmittingIPTrafficoverARCNETNetworks

References
1."History",ARCNETTradeAssociation(http://www.arcnet.com/abtarc.htm#history)
2."ArcnetBoardPartsList".
3."TandyPicksARCNET".
4."ARCNETforControl"(http://www.ccontrols.com/arccontrol/index.htm),ccontrols.com
5."TheRodneyDangerfieldofNetworkComputing"(https://archive.org/stream/ARCNET
DocumentManagement/ARCNETDocumentManagement_djvu.txt),archive.org

Externallinks
ARCNETTradeAssociation(http://www.arcnet.com/)
HistoryofDatapoint,includingARCnet/ARCnetPlusDevelopment
(http://www.fundinguniverse.com/companyhistories/DatapointCorporationCompanyHistory.html)
https://en.wikipedia.org/wiki/ARCNET

4/5

3/26/2016

ARCNETWikipedia,thefreeencyclopedia

Retrievedfrom"https://en.wikipedia.org/w/index.php?title=ARCNET&oldid=693194054"
Categories: Networkprotocols
Thispagewaslastmodifiedon1December2015,at00:31.
TextisavailableundertheCreativeCommonsAttributionShareAlikeLicenseadditionaltermsmay
apply.Byusingthissite,youagreetotheTermsofUseandPrivacyPolicy.Wikipediaisa
registeredtrademarkoftheWikimediaFoundation,Inc.,anonprofitorganization.

https://en.wikipedia.org/wiki/ARCNET

5/5

Common questions

Powered by AI

ARCNET had several key advantages over Ethernet in the early days of local area networking. Firstly, ARCNET employed a star-wired bus topology, which was easier to build, expand and maintain compared to the linear bus topology of Ethernet at the time . This made it easier to add or remove nodes without disrupting the entire network. Secondly, ARCNET could support longer cable runs, with active hubs and coaxial cables enabling connections of up to 610 meters (2,000 feet), compared to the 180 meters (600 feet) limit of Ethernet's RG-58 cable . Additionally, ARCNET's token-passing scheme facilitated a more deterministic and predictable network performance, providing equitable access to the network medium . ARCNET also provided a concrete acknowledgment for message delivery, allowing quicker fault recovery . These attributes made ARCNET more reliable and flexible for certain applications, particularly in environments requiring real-time networking, which contributed to its popularity during its heyday .

ARCNET's impact on embedded systems and industrial control environments has been significant due to its unique strengths. ARCNET's deterministic and token-passing protocol ensured predictable network performance, crucial for real-time control systems where timing precision is necessary . This made it ideal for environments where high reliability and quick recovery from faults are critical, such as industrial automation and process control . Furthermore, its ability to provide a guaranteed maximum time to access the network and equitable medium access without collision delays enabled seamless integration in control systems . Despite being largely replaced by advanced Ethernet technologies in general LAN applications, ARCNET remains valued in industries where these features align with operational needs, ensuring continued niche use .

ARCNET's historical position as an early and popular LAN technology initially gave it a significant market presence. However, its eventual decline was influenced by several market dynamics. Initially, ARCNET was a proprietary solution from Datapoint, which limited broad industry adoption compared to open standards like Ethernet . Although ARCNET was later standardized as ANSI ARCNET 878.1, the momentum had already shifted toward open and faster solutions like Ethernet and its evolving standards . Ethernet's support of higher speeds, along with decreased cost due to industry adoption of twisted pair cabling and technological advancements, made it a more attractive option for enterprises . Furthermore, the competitive response to Ethernet's innovations, such as the move to 10 Mbit/s and later 100 Mbit/s speeds, left ARCNET's incremental innovations like ARCNET Plus unable to recapture market attention . Consequently, despite ARCNET's early dominance and advantages in specific applications, the broader market trends toward faster, more cost-effective, and standardized solutions led to its decline as a LAN protocol .

Datapoint Corporation played a crucial role in the development and commercialization of ARCNET. ARCNET was initially developed by John Murphy, a principal development engineer at Datapoint, in 1976 and announced in 1977 under the leadership of Victor Poor . It was designed to connect Datapoint 2200 terminals to a shared 8" floppy disk system, eventually being repurposed as a local area network (LAN) protocol as microcomputers became prevalent . Datapoint incorporated ARCNET as a reliable and inexpensive LAN solution for microcomputers, paving the way for widespread commercial use by the late 1970s . They also introduced a critical upgrade, ARCNET Plus, in response to Ethernet's competition . However, despite these innovations, Datapoint eventually shifted its focus due to financial troubles .

ARCNET's design and technology were highly suitable for real-time systems due to several key features. The use of a token-passing mechanism instead of Ethernet's CSMA/CD approach allowed for predictable access to the network medium, thus providing deterministic network performance . This was crucial for real-time systems where delay and predictability are critical requirements. Furthermore, each node had to confirm its readiness to receive a message before data transmission, which prevented network time wastage and ensured efficient transfer . ARCNET's ability to guarantee equitable access to the bus and provide concrete acknowledgments for message delivery also meant that error recovery processes were faster . These capabilities made ARCNET an ideal choice for environments where timing and reliability are paramount, such as industrial control and embedded systems .

The development of twisted pair cabling significantly influenced the competition between Ethernet and ARCNET. Previously, Ethernet relied on coaxial cables, which limited its cabling topology and ease of installation. However, with the transition to twisted pair cables, Ethernet adopted an 'interconnected stars' topology using active hubs, making it more attractive due to simpler cabling and installation processes . Furthermore, twisted pair cables supported higher data speeds for Ethernet, providing it with a raw speed advantage (10 Mbit/s) over ARCNET's 2.5 Mbit/s . As Ethernet became faster and the overall costs decreased due to economies of scale, it became more attractive for network installations, leading to a decline in the adoption of ARCNET .

The ARCNET token-passing scheme offered several implications when compared to Ethernet's collision-based protocol. The token-passing method ensured that only one node could transmit at a time, reducing the chances of packet collisions and thus maintaining a stable network performance even under high loads . This predictable and orderly network behavior was particularly beneficial in environments needing real-time data transmission, as it enabled a guaranteed maximum wait time to access the network. In contrast, Ethernet's CSMA/CD approach could lead to performance degradation, especially as the number of devices and traffic increased, because collisions became more frequent, slowing down overall transmission rates . Additionally, the token-passing scheme included built-in mechanisms for acknowledging successful message delivery before passing the token, which enhanced network reliability and fault recovery processes . Therefore, ARCNET was able to maintain robust and predictable performance under conditions that could lead to collapse in an Ethernet network due to excessive collisions .

ARCNET's ability to support longer cable distances was significant in differentiating it from Ethernet, particularly in the early days of LAN technology. This capability allowed ARCNET networks to span greater physical distances with fewer infrastructure installations, thereby reducing the complexity and cost associated with extending network coverage . The maximum run length for ARCNET's coaxial cable between active hubs or between an active hub and a node was 610 meters (2,000 feet), compared to Ethernet's 180 meters (600 feet) limit . This meant that ARCNET was better suited for larger-scale or multi-building deployments without requiring expensive repeaters or additional infrastructure that Ethernet would need to achieve similar coverage. The extended cabling ability also facilitated easier network planning and reduced the overall number of hubs or switches required, which could be a cost-saving factor for installations looking to optimize for distance .

Changes in processor speeds played a pivotal role in affecting the competition between ARCNET and Ethernet. As processor speeds improved, Ethernet's susceptibility to collision-related performance issues diminished significantly. Faster processors were better able to process network collisions and recover swiftly, thereby enhancing Ethernet's performance to match or surpass other protocols . This improvement reduced the earlier advantage of ARCNET's deterministic and collision-free performance, which was more pronounced on slower processors common in earlier computing environments. Consequently, Ethernet's broader adoption was facilitated as its costs also dropped, contributing to the decrease in ARCNET installations as a LAN protocol .

Several factors contributed to the higher cost and limited success of ARCnet Plus upon its introduction. First, by the time ARCnet Plus was introduced, Ethernet had already captured the majority of the network market share due to its higher speed of 10 Mbit/s, compared to ARCnet Plus's 20 Mbit/s, and falling costs due to increased competition . Ethernet's transition from coaxial to twisted pair cabling and the introduction of interconnected star topologies made it easier and more cost-efficient to deploy, further diminishing ARCnet Plus's appeal . Additionally, ARCnet Plus products, primarily produced by Datapoint, were expensive and difficult to source, limiting widespread adoption . Furthermore, industry momentum had shifted toward open networking standards, making proprietary or niche solutions like ARCnet Plus less attractive to consumers . As a result, ARCnet Plus struggled to gain a foothold in a market dominated by rapidly advancing Ethernet technology.

You might also like