138IEEEJOURNALONSELECTEDAREASINCOMMUNICATIONS,VOL.32,NO.1,JANUARY2014
CompressingForwardingTablesfor
DatacenterScalability
OriRottenstreich,MaratRadan,YuvalCassuto,IsaacKeslassy,CarmiArad,TalMizrahi,YoramRevah,andAvinatanHassidim
Abstract—Withtheriseofdatacentervirtualization,thenum-berofentriesintheforwardingtablesofdatacenterswitchesisexpectedtoscalefromseveralthousandstoseveralmillions.Unfortunately,suchforwardingtablesizeswouldnot ton-chipmemoryusingcurrentimplementations.
Inthispaper,weinvestigatethecompressibilityofforwardingtables.We rstintroduceanovelforwardingtablearchitecturewithseparateencodingineachcolumn.Itisdesignedtokeepsupportingfastrandomaccessesand xed-widthmemorywords.Then,weshowthatalthough ndingtheoptimalencodingisNP-hard,wecansuggestanencodingwhosememoryrequirementperrowentryisguaranteedtobewithinasmalladditiveconstantoftheoptimum.Next,weanalyzethecommoncaseoftwo-columnforwardingtables,andshowthatsuchtablescanbepresentedasbipartitegraphs.Wededucegraph-theoreticalboundsontheencodingsize.Wealsointroduceanalgorithmforoptimalconditionalencodingofthesecondcolumngivenanencodingofthe rstone.Inaddition,weexplainhowourarchitecturecanhandletableupdates.Last,weevaluateoursuggestedencodingtechniquesonsyntheticforwardingtablesaswellasonreal-lifetables.
IndexTerms—DatacenterVirtualization,Layer-2Datacenter,ForwardingInformationBase,Compression.
I.INTRODUCTION
A.Background
MACaddressperthread,andconsideringthejointincreaseofthenumbersof(a)threadspercore,(b)coresperprocessor(socket),(c)socketsperserver,and(d)serversperdatacenter,itisclearthattheproductofallthesenumbersisexpectedtodramaticallyincreaseinthecomingyears[1]–[3].Inadditiontothisexpectedstronggrowthinthenumberofentriesperforwardingtable,wecanalsoexpectamoderategrowthinthenumberofattributesperentry,suchasthequality-of-serviceandsecurityattributes,aswellasinthesizeoftheseattributes.Allinall,weneedtobeabletohandledramaticallylargerforwardingtables.
Unfortunately,forwardingtablesmustbeaccessedonaper-packetbasis.Therefore,theyareperformancesensitive.Becauseofthehighdatacenterrates,theyareoftenimple-mentedinhardwareusingon-chipmemory.Thismakesscalingforwardingtableschallenging,becauseon-chipmemoryistoosmalltoholdmillionsofforwarding-tableentriesinrawform[4],[5].
Thegoalofthispaperistostudythecompressibilityofforwardingtables,providingbothananalyticalframeworkandexperimentalinsighttowardtheeffectivecompressionofsuchtables.
B.CompressingForwardingTables
Forwardingtables,alsoreferredtoasForwardingInforma-tionBases(FIBs),consistofseveralentries,whichassociatealookupkeytoitscorrespondingforwardinginformation.Forinstance,thelookupkeycanbea(destinationMACaddress,VLAN)pair,andtheforwardinginformationcanincludetheswitchtargetport,thequality-of-serviceattributes,thesecurityattributes,andadditionalinformationusedforvariousmodi cationsthatneedtobeappliedtothepacket.Forwardingtablesaretypicallyarrangedastwo-dimensionaltables.Eachrowentryinaforwardingtablecorrespondstoadifferentlookupkey,andcontainsseveralcolumns,withthelookupkeyandforwardinginformation.Forwardingtablescanthenbeaccessedusingahashingmechanism.Uponeachincomingpacket,thepacketlookupkeyishashedintoone(orseveral)ofthehashtableentries.Ifthisentryindeedcontainsthelookupkey,thentheforwardingtablereturnstheassociatedforwardinginformation.
Forwardingtablesaretypicallyredundant,becausetheir elds(columns)canhaveahighlynon-uniformvaluedis-tribution,andbecausethevaluesindifferent eldscanbecorrelated.Therefore,weareinterestedinstudyingwhetherthesepropertiescanbeusedtocompressthem.However,we
T
HERAPIDgrowthofforwardingtablesinnetworkswitchesraisesseriousconcernsinthenetworkingindus-try.Layer2(L2)networksarenolongerconstrainedtosmalllocalareanetworks.Onthecontrary,theyarenowusedindatacenternetworks,whichwillsoonhaveaneedformillionsofhostsinasingleL2domain,whilecurrentL2domainsareonlyrestrictedtoseveralthousands[1].Asaresult,currentforwardingtablescannothandlesuchdatacenternetworks.Themaindriverofthisexpecteddramaticgrowthrateisthedeploymentofhostvirtualizationindatacenternet-works.Whereastraditionalserverswouldonlyuseoneortwoglobally-uniqueMACaddressesperserver,contemporaryserversusetensofMACaddresses,correspondingtotheirtensofvirtualmachines(thesegeneratedMACaddressesarenotnecessarilygloballyuniqueanymore).Assumingone
ManuscriptreceivedJanuary15,2013;revisedJuly1,2013.
O.Rottenstreich,M.Radan,Y.CassutoandI.KeslassyarewiththeTechnion,Israel(e-mails:{or@tx,radan@tx,ycassuto@ee,isaac@ee}.technion.ac.il).
C.Arad,T.MizrahiandY.RevaharewithMarvellIsrael(e-mails:{carmi,talmi,yoramr}@marvell.com).T.MizrahiisalsowiththeTechnion.
A.Hassidimiswiththedepartmentofcomputerscience,Bar-IlanUniver-sity,Israel(e-mail:avinatan@cs.biu.ac.il).
DigitalObjectIdenti er10.1109/JSAC.2014.140113.
c2014IEEE0733-8716/14/$31.00
Compressing Video for TaskStream Currently, TaskStream allows users to upload videos up to 100 megabytes in size. Therefore, in order to upload a video...
PPMexe: PPM for Compressing Software Milenko Drini?? and Darko Kirovski? c ? Computer Science Department, UCLA, Los Angeles, USA ? Microsoft Research, ...
Face Fixer Compressing Polygon MeshesFace Fixer Compressing Polygon Meshes隐藏>> UNC Technical Report TR-00-04 Face Fixer: Compressing Polygon Meshes with ...
Compressing Hexahedral Volume Meshes Martin Isenburg? University of North Carolina at Chapel Hill Pierre Alliez? INRIA Sophia-Antipolis Abstract Unstructured ...
Compressing Mixed-State Sources by Sending Classical Information_专业资料。We consider visible compression for discrete memoryless sources of mixed quantum ...
Managing Gigabytes Compressing and Indexing Documents and Images_专业资料。As today’s information explosion generates greater and greater volumes of raw data...
Compressing File Collections with a TSP-Based Approach Dimitre Trendafilov Nasir Memon Torsten Suel Department of Computer and Information Science Technical ...
Compressing cube-connected cycles and butterfly networks We consider the simulation of large cube-connected cycles (CCC) and large butterfly networks (BFN)...
The objectives for compressing DEMs are obtain...更多>>Abstract. Wavelet decomposition is a well-known technique to compress image data. Here, we have ...
Re-Store: A System for Compressing, Browsing, and Searching Large Documents (Extended Abstract of Invited Presentation) Alistair Moffat Raymond Wan ...
热门文档