note Chapter4
NetworkLayer
A on the use of these ppt slides:
Were making these slides freely available to all (faculty, students, readers).
Theyre in PowerPoint form so you can add, modify, and delete slides
(including this one) and slide content to suit your needs. They obviously
represent a lot of work on our part. In return for use, we only ask the
following:
If you use these slides (e.g., in a class) in substantially unaltered form,
that you mention their source (after all, wed like people to use our book!)
If you post any slides in substantially unaltered form on a www site, that
you note that they are adapted from (or perhaps identical to) our slides, and
note our copyright of this material.
ComputerNetworking:ATop
DownApproach
5thedition.
JimKurose,KeithRoss
AddisonWesley,April2009.
Thanks and enjoy! JFK/KWR
All material copyright 1996-2009
J.F Kurose and K.W. Ross, All Rights Reserved
Network Layer
4-1
Chapter4:NetworkLayer
Chaptergoals:
understandprinciplesbehindnetworklayerservices:
r networklayerservicemodels
r forwardingversusrouting
r howarouterworks
r routing(pathselection)
r dealingwithscale
r advancedtopics:IPv6,mobility
instantiation,implementationintheInternet
Network Layer
4-2
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-3
Networklayer
transportsegmentfromsending
toreceivinghost
onsendingsideencapsulates
segmentsintodatagrams
onrcvingside,deliverssegments
totransportlayer
networklayerprotocolsinevery
host,router
routerexaminesheaderfieldsin
allIPdatagramspassing
throughit
application
transport
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
network
datalink
physical
Network Layer
network
datalink
physical
network
datalink
physical
application
transport
network
datalink
physical
4-4
TwoKeyNetworkLayerFunctions
forwarding:movepackets
fromroutersinputto
appropriaterouteroutput
routing:determineroute
analogy:
routing:processofplanning
tripfromsourcetodest
takenbypacketsfromsource forwarding:processof
gettingthroughsingle
todest.
interchange
r
routingalgorithms
Network Layer
4-5
Interplaybetweenroutingandforwarding
routing algorithm
local forwarding table
header
output link
value
0100 3
0101
0111
1001
2
2
1
value in arriving
packets header
0111
1
32
Network Layer
4-6
Connectionsetup
3rdimportantfunctioninsomenetworkarchitectures:
ATM,framerelay,X.25
beforedatagramsflow,twoendhostsandinterveningroutersestablish
virtualconnection
r routersgetinvolved
networkvstransportlayerconnectionservice:
r network:betweentwohosts(mayalsoinvolveinterveningrouters
incaseofVCs)
r transport:betweentwoprocesses
r
Network Layer
4-7
Networkservicemodel
Q:Whatservicemodelforchanneltransportingdatagramsfrom
sendertoreceiver?
Exampleservicesforindividual
datagrams:
guaranteeddelivery
guaranteeddeliverywithless
than40msecdelay
Exampleservicesforaflowof
datagrams:
inorderdatagramdelivery
guaranteedminimum
bandwidthtoflow
restrictionsonchangesin
interpacketspacing
Network Layer
4-8
Networklayerservicemodels:
Network
Architecture
Internet
Service
Model
Guarantees ?
Congestion
Bandwidth Loss Order Timing feedback
best effort none
ATM
CBR
ATM
VBR
ATM
ABR
ATM
UBR
constant
rate
guaranteed
rate
guaranteed
minimum
none
no
no
no
yes
yes
yes
yes
yes
yes
no
yes
no
no (inferred
via loss)
no
congestion
no
congestion
yes
no
yes
no
no
Network Layer
4-9
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
Networklayerconnectionandconnectionless
service
datagramnetworkprovidesnetworklayerconnectionless
service
VCnetworkprovidesnetworklayerconnectionservice
analogoustothetransportlayerservices,but:
service:hosttohost
r nochoice:networkprovidesoneortheother
r implementation:innetworkcore
r
Network Layer
4-
Virtualcircuits
sourcetodestpathbehavesmuchliketelephonecircuit
r
r
performancewise
networkactionsalongsourcetodestpath
callsetup,teardownforeachcallbeforedatacanflow
eachpacketcarriesVCidentifier(notdestinationhostaddress)
everyrouteronsourcedestpathmaintainsstateforeachpassingconnection
link,routerresources(bandwidth,buffers)maybeallocatedtoVC(dedicated
resources=predictableservice)
Network Layer
4-
VCimplementation
aVCconsistsof:
1.
2.
3.
pathfromsourcetodestination
VCnumbers,onenumberforeachlinkalongpath
entriesinforwardingtablesinroutersalongpath
packetbelongingtoVCcarriesVCnumber(ratherthan
destaddress)
VCnumbercanbechangedoneachlink.
r
NewVCnumbercomesfromforwardingtable
Network Layer
4-
Forwardingtable
VCnumber
22
12
1
Forwardingtablein
northwestrouter:
2
32
3
interface
number
IncominginterfaceIncomingVC#OutgoinginterfaceOutgoingVC#
112322
263118
37217
197387
Routersmaintainconnectionstateinformation!
Network Layer
4-
Virtualcircuits:signalingprotocols
usedtosetup,maintainteardownVC
usedinATM,framerelay,X.25
notusedintodaysInternet
application
transport
network
datalink
physical
5.Dataflowbegins
4.Callconnected
1.Initiatecall
6.Receivedata
3.Acceptcall
2.incomingcall
Network Layer
application
transport
network
datalink
physical
4-
Datagramnetworks
nocallsetupatnetworklayer
routers:nostateaboutendtoendconnections
r nonetworklevelconceptofconnection
packetsforwardedusingdestinationhostaddress
r packetsbetweensamesourcedestpairmaytakedifferentpaths
application
transport
network
datalink
physical
1.Senddata
2.Receivedata
Network Layer
application
transport
network
datalink
physical
4-
Forwardingtable
Destination Address Range
4billion
possibleentries
Link Interface
11001000 00010111 00010000 00000000
through
11001000 00010111 00010111 11111111
0
11001000 00010111 00011000 00000000
through
11001000 00010111 00011000 11111111
1
11001000 00010111 00011001 00000000
through
11001000 00010111 00011111 11111111
2
otherwise
3
Network Layer
4-
Longestprefixmatching
Prefix Match
11001000 00010111 00010
11001000 00010111 00011000
11001000 00010111 00011
otherwise
Link Interface
0
1
2
3
Examples
Whichinterface?
DA:11001000000101110001011010100001
DA:11001000000101110001100010101010
Whichinterface?
Network Layer
4-
DatagramorVCnetwork:why?
Internet(datagram)
dataexchangeamongcomputers
ATM(VC)
evolvedfromtelephony
elasticservice,nostricttiming humanconversation:
req.
r stricttiming,reliability
smartendsystems(computers)
requirements
r canadapt,performcontrol,error
r needforguaranteedservice
recovery
dumbendsystems
r simpleinsidenetwork,complexity
r telephones
atedge
r complexityinsidenetwork
manylinktypes
r differentcharacteristics
r uniformservicedifficult
r
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
RouterArchitectureOverview
Twokeyrouterfunctions:
runroutingalgorithms/protocol(RIP,OSPF,BGP)
forwardingdatagramsfromincomingtooutgoinglink
Network Layer
4-
InputPortFunctions
Physicallayer:
bitlevelreception
Datalinklayer:
e.g.,Ethernet
seechapter5
Decentralized switching:
givendatagramdest.,lookupoutputportusing
forwardingtableininputportmemory
goal:completeinputportprocessingatlinespeed
queuing:ifdatagramsarrivefasterthanforwarding
rateintoswitchfabric
Network Layer
4-
Threetypesofswitchingfabrics
Network Layer
4-
SwitchingViaMemory
Firstgenerationrouters:
traditionalcomputerswithswitchingunderdirectcontrolofCPU
packetcopiedtosystemsmemory
speedlimitedbymemorybandwidth(2buscrossingsperdatagram)
Input
Port
Memory
Output
Port
System Bus
Network Layer
4-
SwitchingViaaBus
datagramfrominputportmemory
tooutputportmemoryviaasharedbus
buscontention:switchingspeedlimitedbybus
bandwidth
32Gbpsbus,Cisco5600:sufficientspeedfor
accessandenterpriserouters
Network Layer
4-
SwitchingViaAnInterconnectionNetwork
overcomebusbandwidthlimitations
Banyannetworks,otherinterconnectionnetsinitiallydevelopedto
connectprocessorsinmultiprocessor
advanceddesign:fragmentingdatagramintofixedlengthcells,
switchcellsthroughthefabric.
Cisco12000:switches60Gbpsthroughtheinterconnection
network
Network Layer
4-
OutputPorts
Bufferingrequiredwhendatagramsarrivefromfabricfasterthan
thetransmissionrate
Schedulingdisciplinechoosesamongqueueddatagramsfor
transmission
Network Layer
4-
Outputportqueueing
bufferingwhenarrivalrateviaswitchexceedsoutputlinespeed
queueing(delay)andlossduetooutputportbufferoverflow!
Network Layer
4-
Howmuchbuffering?
RFC3439ruleofthumb:averagebufferingequalto
typicalRTT(say250msec)timeslinkcapacityC
r
e.g.,C=10Gpslink:2.5Gbitbuffer
Recentrecommendation:withNflows,bufferingequalto
.
RTTC
N
Network Layer
4-
InputPortQueuing
Fabricslowerthaninputportscombined>queueingmayoccurat
inputqueues
HeadoftheLine(HOL)blocking:queueddatagramatfrontofqueue
preventsothersinqueuefrommovingforward
queueingdelayandlossduetoinputbufferoverflow!
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
TheInternetNetworklayer
Host,routernetworklayerfunctions:
Transportlayer:TCP,UDP
Network
layer
IPprotocol
addressingconventions
datagramformat
packethandlingconventions
Routingprotocols
pathselection
RIP,OSPF,BGP
forwarding
table
ICMPprotocol
errorreporting
routersignaling
Linklayer
physicallayer
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
IPdatagramformat
IPprotocolversion
number
headerlength
(bytes)
typeofdata
maxnumber
remaininghops
(decrementedat
eachrouter)
upperlayerprotocol
todeliverpayloadto
howmuchoverheadwith
TCP?
20bytesofTCP
20bytesofIP
=40bytes+applayer
overhead
32bits
ver head.
len
typeof
service
16bitidentifier
upper
timeto
layer
live
length
fragment
offset
header
checksum
flgs
totaldatagram
length(bytes)
for
fragmentation/
reassembly
32bitsourceIPaddress
32bitdestinationIPaddress
Options(ifany)
data
(variablelength,
typicallyaTCP
orUDPsegment)
Network Layer
E.g.timestamp,
recordroute
taken,specify
listofrouters
tovisit.
4-
IPFragmentation&Reassembly
networklinkshaveMTU(max.transfer
size)largestpossiblelinklevelframe.
r differentlinktypes,different
MTUs
largeIPdatagramdivided
(fragmented)withinnet
r onedatagrambecomesseveral
datagrams
r reassembledonlyatfinal
destination
r IPheaderbitsusedtoidentify,
orderrelatedfragments
fragmentation:
in:onelargedatagram
out:3smallerdatagrams
reassembly
Network Layer
4-
IPFragmentationandReassembly
Example
4000bytedatagram
MTU=1500bytes
length ID
=4000 =x
fragflag
=0
offset
=0
Onelargedatagrambecomes
severalsmallerdatagrams
length ID
=1500 =x
1480bytesin
datafield
offset=
1480/8
fragflag
=1
offset
=0
length ID
=1500 =x
fragflag
=1
offset
=185
length ID
=1040 =x
fragflag
=0
offset
=370
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
IPAddressing:introduction
IPaddress:32bitidentifier
forhost,routerinterface
interface:connection
betweenhost/routerand
physicallink
r
r
r
routerstypicallyhave
multipleinterfaces
hosttypicallyhasoneinterface
IPaddressesassociatedwith
eachinterface
223.1.1.1
223.1.1.2
223.1.1.4
223.1.1.3
223.1.2.1
223.1.2.9
223.1.3.27
223.1.2.2
223.1.3.2
223.1.3.1
223.1.1.1 = 11011111 00000001 00000001 00000001
223
1
1
Network Layer
1
4-
Subnets
IPaddress:
r subnetpart(highorderbits)
r hostpart(loworderbits)
Whatsasubnet?
r deviceinterfaceswithsame
subnetpartofIPaddress
r canphysicallyreacheachother
withoutinterveningrouter
223.1.1.1
223.1.1.2
223.1.1.4
223.1.1.3
223.1.2.1
223.1.2.9
223.1.3.27
223.1.2.2
subnet
223.1.3.1
223.1.3.2
networkconsistingof3subnets
Network Layer
4-
Subnets
223.1.1.0/24
223.1.2.0/24
Recipe
Todeterminethesubnets,
detacheachinterfacefromits
hostorrouter,creatingislands
ofisolatednetworks.Each
isolatednetworkiscalleda
subnet.
223.1.3.0/24
Subnetmask:/24
Network Layer
4-
Subnets
223.1.1.2
Howmany?
223.1.1.1
223.1.1.4
223.1.1.3
223.1.7.0
223.1.9.2
223.1.9.1
223.1.7.1
223.1.8.1
223.1.8.0
223.1.2.6
223.1.2.1
223.1.3.27
223.1.2.2
223.1.3.1
223.1.3.2
Network Layer
4-
IPaddressing:CIDR
CIDR:ClasslessInterDomainRouting
subnetportionofaddressofarbitrarylength
r addressformat:a.b.c.d/x,wherexis#bitsinsubnetportionof
address
r
subnet
part
host
part
11001000 00010111 00010000 00000000
200.23.16.0/23
Network Layer
4-
IPaddresses:howtogetone?
Q:HowdoesahostgetIPaddress?
hardcodedbysystemadmininafile
Windows:controlpanel>network>configuration>tcp/ip
>properties
r UNIX:/etc/rc.config
DHCP:DynamicHostConfigurationProtocol:dynamicallyget
addressfromasserver
r plugandplay
r
Network Layer
4-
DHCP:DynamicHostConfigurationProtocol
Goal:allowhosttodynamicallyobtainitsIPaddressfromnetworkserverwhen
itjoinsnetwork
Canrenewitsleaseonaddressinuse
Allowsreuseofaddresses(onlyholdaddresswhileconnectedanon)
Supportformobileuserswhowanttojoinnetwork(moreshortly)
DHCPoverview:
r hostbroadcastsDHCPdiscovermsg[optional]
r DHCPserverrespondswithDHCPoffermsg[optional]
r hostrequestsIPaddress:DHCPrequestmsg
r DHCPserversendsaddress:DHCPackmsg
Network Layer
4-
DHCPclientserverscenario
A
B
223.1.1.2
223.1.1.4
223.1.1.3
223.1.3.1
223.1.2.1
DHCP
server
223.1.1.1
223.1.2.9
223.1.3.27
223.1.2.2
223.1.3.2
E
arriving DHCP
client needs
address in this
network
Network Layer
4-
DHCPclientserverscenario
DHCPserver:223.1.2.5
DHCP discover
src : 0.0.0.0, 68
dest.: 255.255.255.255,67
yiaddr: 0.0.0.0
transaction ID: 654
arriving
client
DHCP offer
src: 223.1.2.5, 67
dest: 255.255.255.255, 68
yiaddrr: 223.1.2.4
transaction ID: 654
Lifetime: 3600 secs
DHCP request
time
src: 0.0.0.0, 68
dest:: 255.255.255.255, 67
yiaddrr: 223.1.2.4
transaction ID: 655
Lifetime: 3600 secs
DHCP ACK
src: 223.1.2.5, 67
dest: 255.255.255.255, 68
yiaddrr: 223.1.2.4
transaction ID: 655
Lifetime: 3600 secs
Network Layer
4-
DHCP:morethanIPaddress
DHCPcanreturnmorethanjustallocatedIPaddresson
subnet:
addressoffirsthoprouterforclient
r nameandIPaddressofDNSsever
r networkmask(indicatingnetworkversushostportionof
address)
r
Network Layer
4-
DHCP:example
DHCP
UDP
IP
Eth
Phy
DHCP
DHCP
DHCP
DHCP
DHCPrequestencapsulatedin
DHCP
DHCP
DHCP
DHCP
DHCP
connectinglaptopneedsitsIP
address,addroffirsthoprouter,
addrofDNSserver:useDHCP
DHCP
UDP
IP
Eth
Phy
168.1.1.1
router
(runsDHCP)
UDP,encapsulatedinIP,
encapsulatedin802.1Ethernet
Ethernetframebroadcast(dest:
FFFFFFFFFFFF)onLAN,receivedat
routerrunningDHCPserver
EthernetdemuxedtoIPdemuxed,
UDPdemuxedtoDHCP
Network Layer
4-
DHCP:example
DHCP
UDP
IP
Eth
Phy
DHCP
DHCP
DHCP
DHCP
DCPserverformulatesDHCPACK
containingclientsIPaddress,IP
addressoffirsthoprouterfor
client,name&IPaddressofDNS
server
encapsulationofDHCPserver,
DHCP
DHCP
DHCP
DHCP
DHCP
DHCP
UDP
IP
Eth
Phy
router
(runsDHCP)
frameforwardedtoclient,
demuxinguptoDHCPatclient
clientnowknowsitsIPaddress,
nameandIPaddressofDSN
server,IPaddressofitsfirsthop
router
Network Layer
4-
DHCP:wireshark
output(homeLAN)
Message type: Boot Request (1)
Hardware type: Ethernet
Hardware address length: 6
Hops: 0
Transaction ID: 0x6b3a11b7
Seconds elapsed: 0
Bootp flags: 0x0000 (Unicast)
Client IP address: 0.0.0.0 (0.0.0.0)
Your (client) IP address: 0.0.0.0 (0.0.0.0)
Next server IP address: 0.0.0.0 (0.0.0.0)
Relay agent IP address: 0.0.0.0 (0.0.0.0)
Client MAC address: Wistron_23:68:8a (00:16:d3:23:68:8a)
Server host name not given
Boot file name not given
Magic cookie: (OK)
Option: (t=53,l=1) DHCP Message Type = DHCP Request
Option: (61) Client identifier
Length: 7; Value: 010016D323688A;
Hardware type: Ethernet
Client MAC address: Wistron_23:68:8a (00:16:d3:23:68:8a)
Option: (t=50,l=4) Requested IP Address = 192.168.1.101
Option: (t=12,l=5) Host Name = "nomad"
Option: (55) Parameter Request List
Length: 11; Value: 010F03062C2E2F1F21F92B
1 = Subnet Mask; 15 = Domain Name
3 = Router; 6 = Domain Name Server
44 = NetBIOS over TCP/IP Name Server
request
reply
Message type: Boot Reply (2)
Hardware type: Ethernet
Hardware address length: 6
Hops: 0
Transaction ID: 0x6b3a11b7
Seconds elapsed: 0
Bootp flags: 0x0000 (Unicast)
Client IP address: 192.168.1.101 (192.168.1.101)
Your (client) IP address: 0.0.0.0 (0.0.0.0)
Next server IP address: 192.168.1.1 (192.168.1.1)
Relay agent IP address: 0.0.0.0 (0.0.0.0)
Client MAC address: Wistron_23:68:8a (00:16:d3:23:68:8a)
Server host name not given
Boot file name not given
Magic cookie: (OK)
Option: (t=53,l=1) DHCP Message Type = DHCP ACK
Option: (t=54,l=4) Server Identifier = 192.168.1.1
Option: (t=1,l=4) Subnet Mask = 255.255.255.0
Option: (t=3,l=4) Router = 192.168.1.1
Option: (6) Domain Name Server
Length: 12; Value: 445747E2445749F244574092;
IP Address: 68.87.71.226;
IP Address: 68.87.73.242;
IP Address: 68.87.64.146
Option: (t=15,l=20) Domain Name = "hsd1.ma.comcast.net."
Network Layer
4-
IPaddresses:howtogetone?
Q:HowdoesnetworkgetsubnetpartofIPaddr?
A:getsallocatedportionofitsproviderISPsaddressspace
ISP's block
11001000 00010111 00010000 00000000
200.23.16.0/20
Organization 0
Organization 1
Organization 2
...
11001000 00010111 00010000 00000000
11001000 00010111 00010010 00000000
11001000 00010111 00010100 00000000
..
.
200.23.16.0/23
200.23.18.0/23
200.23.20.0/23
.
Organization 7
11001000 00010111 00011110 00000000
200.23.30.0/23
Network Layer
4-
Hierarchicaladdressing:routeaggregation
Hie ra rc h ic a la d d re s s in g a llo ws e ffic ie nta d ve rtis e m e nto fro uting
info rm a tio n:
Organization0
200.23.16.0/23
Organization1
200.23.18.0/23
Organization2
200.23.20.0/23
Organization7
.
.
.
.
.
.
FlyByNightISP
Sendmeanything
withaddresses
beginning
200.23.16.0/20
Internet
200.23.30.0/23
ISPsRUs
Sendmeanything
withaddresses
beginning
199.31.0.0/16
Network Layer
4-
Hierarchicaladdressing:morespecificroutes
ISPsRUshasamorespecificroutetoOrganization1
Organization0
200.23.16.0/23
Organization2
200.23.20.0/23
Organization7
.
.
.
.
.
.
FlyByNightISP
Sendmeanything
withaddresses
beginning
200.23.16.0/20
Internet
200.23.30.0/23
ISPsRUs
Organization1
200.23.18.0/23
Sendmeanything
withaddresses
beginning199.31.0.0/16
or200.23.18.0/23
Network Layer
4-
IPaddressing:thelastword...
Q:HowdoesanISPgetblockofaddresses?
A:ICANN:InternetCorporationforAssigned
NamesandNumbers
r allocatesaddresses
r managesDNS
r assignsdomainnames,resolvesdisputes
Network Layer
4-
NAT:NetworkAddressTranslation
restof
Internet
localnetwork
(e.g.,homenetwork)
10.0.0/24
10.0.0.4
10.0.0.1
10.0.0.2
138.76.29.7
10.0.0.3
Alldatagramsleavinglocal
networkhavesamesinglesourceNATIP
address:138.76.29.7,
differentsourceportnumbers
Datagramswithsourceor
destinationinthisnetwork
have10.0.0/24addressfor
source,destination(asusual)
Network Layer
4-
NAT:NetworkAddressTranslation
Motivation:localnetworkusesjustoneIPaddressasfarasoutsideworld
isconcerned:
r rangeofaddressesnotneededfromISP:justoneIPaddressforall
devices
r canchangeaddressesofdevicesinlocalnetworkwithoutnotifying
outsideworld
r canchangeISPwithoutchangingaddressesofdevicesinlocalnetwork
r devicesinsidelocalnetnotexplicitlyaddressable,visiblebyoutside
world(asecurityplus).
Network Layer
4-
NAT:NetworkAddressTranslation
Implementation:NATroutermust:
r
outgoingdatagrams:replace(sourceIPaddress,port#)ofevery
outgoingdatagramto(NATIPaddress,newport#)
...remoteclients/serverswillrespondusing(NATIPaddress,new
port#)asdestinationaddr.
r
remember(inNATtranslationtable)every(sourceIPaddress,port#)
to(NATIPaddress,newport#)translationpair
r
incomingdatagrams:replace(NATIPaddress,newport#)indest
fieldsofeveryincomingdatagramwithcorresponding(sourceIP
address,port#)storedinNATtable
Network Layer
4-
NAT:NetworkAddressTranslation
NATtranslationtable
WANsideaddrLANsideaddr
2:NATrouter
changesdatagram
sourceaddrfrom
10.0.0.1,3345to
138.76.29.7,5001,
updatestable
1:host10.0.0.1
sendsdatagramto
128.119.40.186,80
138.76.29.7,500110.0.0.1,3345
S:10.0.0.1,3345
D:128.119.40.186,80
2
1
S:138.76.29.7,5001
D:128.119.40.186,80
10.0.0.4
138.76.29.7
S:128.119.40.186,80
D:138.76.29.7,5001
3:Replyarrives
dest.address:
138.76.29.7,5001
3
10.0.0.1
10.0.0.2
S:128.119.40.186,80
D:10.0.0.1,3345
4
10.0.0.3
4:NATrouter
changesdatagram
destaddrfrom
138.76.29.7,5001to10.0.0.1,3345
Network Layer
4-
NAT:NetworkAddressTranslation
16bitportnumberfield:
r 60,000simultaneousconnectionswithasingleLANside
address!
NATiscontroversial:
r routersshouldonlyprocessuptolayer3
r violatesendtoendargument
NATpossibilitymustbetakenintoaccountbyappdesigners,eg,P2P
applications
r
addressshortageshouldinsteadbesolvedbyIPv6
Network Layer
4-
NATtraversalproblem
clientwantstoconnecttoserver
withaddress10.0.0.1
r
r
serveraddress10.0.0.1localtoLAN
(clientcantuseitasdestinationaddr)
onlyoneexternallyvisibleNATted
address:138.76.29.7
solution1:staticallyconfigureNAT
toforwardincomingconnection
requestsatgivenporttoserver
r
Client
10.0.0.1
?
10.0.0.4
138.76.29.7
NAT
router
e.g.,(123.76.29.7,port2500)always
forwardedto10.0.0.1port25000
Network Layer
4-
NATtraversalproblem
solution2:UniversalPlugandPlay
(UPnP)InternetGatewayDevice(IGD)
Protocol.AllowsNATtedhostto:
learnpublicIPaddress(138.76.29.7)
add/removeportmappings(withlease
times)
i.e.,automatestaticNATportmap
configuration
10.0.0.1
IGD
10.0.0.4
138.76.29.7
NAT
router
Network Layer
4-
NATtraversalproblem
solution3:relaying(usedinSkype)
NATedclientestablishesconnectiontorelay
r Externalclientconnectstorelay
r relaybridgespacketsbetweentoconnections
r
2.connectionto
relayinitiated
byclient
Client
3.relaying
established
1.connectionto
relayinitiated
byNATtedhost
138.76.29.7
10.0.0.1
NAT
router
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
ICMP:InternetControlMessageProtocol
usedbyhosts&routerstocommunicate
networklevelinformation
r errorreporting:unreachablehost,
network,port,protocol
r echorequest/reply(usedbyping)
networklayeraboveIP:
r ICMPmsgscarriedinIP
datagrams
ICMPmessage:type,codeplusfirst8
bytesofIPdatagramcausingerror
Type
0
3
3
3
3
3
3
4
Code
0
0
1
2
3
6
7
0
8
9
10
11
12
0
0
0
0
0
description
echo reply (ping)
dest. network unreachable
dest host unreachable
dest protocol unreachable
dest port unreachable
dest network unknown
dest host unknown
source quench (congestion
control - not used)
echo request (ping)
route advertisement
router discovery
TTL expired
bad IP header
Network Layer
4-
TracerouteandICMP
SourcesendsseriesofUDP
segmentstodest
r
r
r
FirsthasTTL=1
SecondhasTTL=2,etc.
Unlikelyportnumber
Whennthdatagramarrivestonth
router:
r
r
r
Routerdiscardsdatagram
AndsendstosourceanICMP
message(type11,code0)
Messageincludesnameofrouter&
IPaddress
WhenICMPmessagearrives,source
calculatesRTT
Traceroutedoesthis3times
Stoppingcriterion
UDPsegmenteventuallyarrivesat
destinationhost
DestinationreturnsICMPhost
unreachablepacket(type3,code3)
WhensourcegetsthisICMP,stops.
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
IPv6
Initialmotivation:32bitaddressspacesoontobecompletely
allocated.
Additionalmotivation:
headerformathelpsspeedprocessing/forwarding
r headerchangestofacilitateQoS
IPv6datagramformat:
r fixedlength40byteheader
r nofragmentationallowed
r
Network Layer
4-
IPv6Header(Cont)
Priority:identifypriorityamongdatagramsinflow
FlowLabel:identifydatagramsinsameflow.
(conceptofflownotwelldefined).
Nextheader:identifyupperlayerprotocolfordata
Network Layer
4-
OtherChangesfromIPv4
Checksum:removedentirelytoreduceprocessingtimeat
eachhop
Options:allowed,butoutsideofheader,indicatedby
NextHeaderfield
ICMPv6:newversionofICMP
additionalmessagetypes,e.g.PacketTooBig
r multicastgroupmanagementfunctions
r
Network Layer
4-
TransitionFromIPv4ToIPv6
Notallrouterscanbeupgradedsimultaneous
r HowwillthenetworkoperatewithmixedIPv4andIPv6routers?
noflagdays
r Tunneling:IPv6carriedaspayloadinIPv4datagramamong
IPv4routers
Network Layer
4-
Tunneling
E
F
IPv6
IPv6
IPv6
A
B
E
F
Physicalview:
IPv6
IPv6
IPv6
IPv6
B
IPv6
Logicalview:
A
tunnel
IPv4
IPv4
Network Layer
4-
Tunneling
Physicalview:
B
C
IPv6
IPv6
IPv4
F
IPv6
D
E
F
IPv4
IPv6
IPv6
IPv6
A
E
IPv6
B
IPv6
Logicalview:
A
tunnel
data
AtoB:
IPv6
Src:B
Dest:E
Src:B
Dest:E
Flow:X
Src:A
Dest:F
Flow:X
Src:A
Dest:F
data
Flow:X
Src:A
Dest:F
data
BtoC:
IPv6inside
IPv4
BtoC:
IPv6inside
IPv4
Flow:X
Src:A
Dest:F
data
EtoF:
IPv6
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
Interplaybetweenrouting,forwarding
routing algorithm
local forwarding table
header
output link
value
0100 3
0101
0111
1001
2
2
1
value in arriving
packets header
0111
1
32
Network Layer
4-
Graphabstraction
5
2
u
1
Graph: G = (N,E)
3
v
2
x
w
3
1
5
z
1
y
2
N = set of routers = { u, v, w, x, y, z }
E = set of links ={ (u,v), (u,x), (v,x), (v,w), (x,w), (x,y), (w,y), (w,z), (y,z) }
Remark:Graphabstractionisusefulinothernetworkcontexts
Example:P2P,whereNissetofpeersandEissetofTCPconnections
Network Layer
4-
Graphabstraction:costs
5
2
u
1
c(x,x)=costoflink(x,x)
3
v
2
x
w
3
1
5
z
1
y
2
e.g.,c(w,z)=5
costcouldalwaysbe1,or
inverselyrelatedtobandwidth,
orinverselyrelatedto
congestion
Costofpath(x1,x2,x3,,xp)=c(x1,x2)+c(x2,x3)++c(xp1,xp)
Question:Whatstheleastcostpathbetweenuandz?
Routingalgorithm:algorithmthatfindsleastcostpath
Network Layer
4-
RoutingAlgorithmclassification
Globalordecentralized
information?
Global:
allroutershavecompletetopology,
linkcostinfo
linkstatealgorithms
Decentralized:
routerknowsphysicallyconnected
neighbors,linkcoststoneighbors
iterativeprocessofcomputation,
exchangeofinfowithneighbors
distancevectoralgorithms
Staticordynamic?
Static:
routeschangeslowlyovertime
Dynamic:
routeschangemorequickly
r periodicupdate
r inresponsetolinkcost
changes
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
ALinkStateRoutingAlgorithm
Dijkstrasalgorithm
nettopology,linkcostsknowntoall
nodes
r accomplishedvialinkstate
broadcast
r allnodeshavesameinfo
computesleastcostpathsfromone
node(source)toallothernodes
r givesforwardingtableforthat
node
iterative:afterkiterations,know
leastcostpathtokdest.s
Notation:
c(x,y):linkcostfromnodextoy;
=ifnotdirectneighbors
D(v):currentvalueofcostofpath
fromsourcetodest.v
p(v):predecessornodealongpath
fromsourcetov
N':setofnodeswhoseleastcost
pathdefinitivelyknown
Network Layer
4-
DijsktrasAlgorithm
1 Initialization:
2 N' = {u}
3 for all nodes v
4
if v adjacent to u
5
then D(v) = c(u,v)
6
else D(v) =
7
8 Loop
9 find w not in N' such that D(w) is a minimum
10 add w to N'
11 update D(v) for all v adjacent to w and not in N ' :
12
D(v) = min( D(v), D(w) + c(w,v) )
13 /* new cost to v is either old cost to v or known
14 shortest path cost to w plus cost from w to v */
15 until all nodes in N'
Network Layer
4-
Dijkstrasalgorithm:example
Step
0
1
2
3
4
5
N'
u
ux
uxy
uxyv
uxyvw
uxyvwz
D(v),p(v) D(w),p(w)
2,u
5,u
2,u
4,x
2,u
3,y
3,y
D(x),p(x)
1,u
D(y),p(y)
2,x
D(z),p(z)
4,y
4,y
4,y
5
2
u
1
3
v
2
x
w
3
1
5
z
1
y
2
Network Layer
4-
Dijkstrasalgorithm:example(2)
Resultingshortestpathtreefromu:
v
w
u
z
x
y
Resultingforwardingtableinu:
destination
link
v
x
(u,v)
(u,x)
y
(u,x)
w
(u,x)
z
(u,x)
Network Layer
4-
Dijkstrasalgorithm,discussion
Algorithmcomplexity:nnodes
eachiteration:needtocheckallnodes,w,notinN
n(n+1)/2comparisons:O(n2)
moreefficientimplementationspossible:O(nlogn)
Oscillationspossible:
e.g.,linkcost=amountofcarriedtraffic
D
1
A
1
0
0
e
0
C
1+e
e
initially
B
1
2+e
D
0
A
1+e 1
C
0
0
B
recompute
routing
0
D
1
A
00
C
2+e
1+e
recompute
B
2+e
D
0
A
1+e 1
C
0
e
B
recompute
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
DistanceVectorAlgorithm
BellmanFordEquation(dynamicprogramming)
Define
dx(y):=costofleastcostpathfromxtoy
Then
dx(y)=min{c(x,v)+dv(y)}
v
whereministakenoverallneighborsvofx
Network Layer
4-
BellmanFordexample
5
2
u
1
3
v
2
w
3
Clearly,dv(z)=5,dx(z)=3,dw(z)=3
5
z
1
BFequationsays:
du(z)=min{c(u,v)+dv(z),
y
1
c(u,x)+dx(z),
c(u,w)+dw(z)}
=min{2+5,
1+3,
5+3}=4
Nodethatachievesminimumisnext
hopinshortestpath forwardingtable
x
2
Network Layer
4-
DistanceVectorAlgorithm
Dx(y)=estimateofleastcostfromxtoy
Nodexknowscosttoeachneighborv:c(x,v)
NodexmaintainsdistancevectorDx=[Dx(y):yN]
Nodexalsomaintainsitsneighborsdistancevectors
r Foreachneighborv,xmaintains
Dv=[Dv(y):yN]
Network Layer
4-
Distancevectoralgorithm(4)
Basicidea:
Fromtimetotime,eachnodesendsitsowndistancevector
estimatetoneighbors
Asynchronous
WhenanodexreceivesnewDVestimatefromneighbor,itupdates
itsownDVusingBFequation:
Dx(y)minv{c(x,v)+Dv(y)}foreachnodeyN
Underminor,naturalconditions,theestimateDx(y)convergetothe
actualleastcost dx(y)
Network Layer
4-
DistanceVectorAlgorithm(5)
Iterative,asynchronous:eachlocal
iterationcausedby:
locallinkcostchange
DVupdatemessagefromneighbor
Distributed:
eachnodenotifiesneighborsonly
whenitsDVchanges
r
neighborsthennotifytheir
neighborsifnecessary
Eachnode:
wait for (change in local link
cost or msg from neighbor)
recompute estimates
if DV to any dest has
changed, notify neighbors
Network Layer
4-
from
x 027
y
z
node y table
costto
xyz
costto
xyz
from
node x table
costto
xyz
Dx(z) = min{c(x,y) +
Dy(z),c(x,z)+Dz(z)}
=min{2+1,7+0}=3
Dx(y)=min{c(x,y)+Dy(y),c(x,z)+Dz(y)}
=min{2+0,7+1}=2
x
y
z
0 2 3
201
710
x
y 201
z
node z table
costto
xyz
from
from
x
x
y
z 71 0
time
2
Network Layer
y
7
1
z
4-
x
y
z 71 0
201
710
from
from
023
x
y
z
027
201
710
from
x
y
z
x
y
z
costto
xyz
x
y
z
027
201
310
023
201
310
costto
xyz
023
201
310
x
2
y
7
1
z
costto
xyz
from
from
from
x
y 201
z
node z table
costto
xyz
x
y
z
costto
xyz
costto
xyz
from
from
x 027
y
z
node y table
costto
xyz
costto
xyz
from
node x table
costto
xyz
Dx(z) = min{c(x,y) +
Dy(z),c(x,z)+Dz(z)}
=min{2+1,7+0}=3
Dx(y)=min{c(x,y)+Dy(y),c(x,z)+Dz(y)}
=min{2+0,7+1}=2
x
y
z
023
201
310
time
Network Layer
4-
DistanceVector:linkcostchanges
Linkcostchanges:
nodedetectslocallinkcostchange
updatesroutinginfo,recalculates
distancevector
ifDVchanges,notifyneighbors
good
news
travels
fast
1
x
4
y
1
50
z
Attimet0,ydetectsthelinkcostchange,updatesitsDV,
andinformsitsneighbors.
Attimet1,zreceivestheupdatefromyandupdatesitstable.
ItcomputesanewleastcosttoxandsendsitsneighborsitsDV.
Attimet2,yreceiveszsupdateandupdatesitsdistancetable.
ysleastcostsdonotchangeandhenceydoesnotsendany
messagetoz.
Network Layer
4-
DistanceVector:linkcostchanges
Linkcostchanges:
goodnewstravelsfast
badnewstravelsslowcountto
infinityproblem!
44iterationsbeforealgorithm
stabilizes:seetext
60
x
4
y
50
1
z
Poisonedreverse:
IfZroutesthroughYtogettoX:
r
ZtellsYits(Zs)distancetoXis
infinite(soYwontroutetoXvia
Z)
willthiscompletelysolvecountto
infinityproblem?
Network Layer
4-
ComparisonofLSandDValgorithms
Messagecomplexity
LS:withnnodes,Elinks,O(nE)msgs
sent
DV:exchangebetweenneighborsonly
r convergencetimevaries
SpeedofConvergence
LS:O(n2)algorithmrequiresO(nE)
msgs
r mayhaveoscillations
DV:convergencetimevaries
r mayberoutingloops
r counttoinfinityproblem
Robustness:whathappensifrouter
malfunctions?
LS:
r
r
DV:
r
r
nodecanadvertiseincorrectlink
cost
eachnodecomputesonlyitsown
table
DVnodecanadvertiseincorrect
pathcost
eachnodestableusedbyothers
errorpropagatethrunetwork
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
HierarchicalRouting
Ourroutingstudythusfaridealization
allroutersidentical
networkflat
nottrueinpractice
scale:with200million
destinations:
cantstorealldestsinrouting
tables!
routingtableexchangewould
swamplinks!
administrativeautonomy
internet=networkofnetworks
eachnetworkadminmaywantto
controlroutinginitsownnetwork
Network Layer
4-
HierarchicalRouting
aggregateroutersintoregions,
autonomoussystems(AS)
routersinsameASrunsame
routingprotocol
r
r
Gatewayrouter
Directlinktorouterin
anotherAS
intraASroutingprotocol
routersindifferentAScanrun
differentintraASrouting
protocol
Network Layer
4-
InterconnectedASes
3c
3b
3a
AS3
1a
2a
1c
1d
1b
Intra-AS
Routing
algorithm
2c
AS2
AS1
Inter-AS
Routing
algorithm
Forwarding
table
2b
forwardingtableconfigured
bybothintraandinterAS
routingalgorithm
r
r
intraASsetsentriesfor
internaldests
interAS&intraAssetsentries
forexternaldests
Network Layer
4-
InterAStasks
AS1must:
1. learnwhichdestsare
reachablethroughAS2,
whichthroughAS3
2. propagatethisreachability
infotoallroutersinAS1
JobofinterASrouting!
supposerouterinAS1receives
datagramdestinedoutsideof
AS1:
r routershouldforward
packettogatewayrouter,
butwhichone?
3c
3b
3a
AS3
1a
2a
1c
1d
1b
2c
AS2
2b
AS1
Network Layer
4-
Example:Settingforwardingtableinrouter1d
supposeAS1learns(viainterASprotocol)thatsubnetxreachablevia
AS3(gateway1c)butnotviaAS2.
interASprotocolpropagatesreachabilityinfotoallinternalrouters.
router1ddeterminesfromintraASroutinginfothatitsinterface Iison
theleastcostpathto1c.
r installsforwardingtableentry(x,I)
3c
3a
3b
AS3
1a
x
2a
1c
1d
1b AS1
2c
AS2
2b
Network Layer
4-
Example:ChoosingamongmultipleASes
nowsupposeAS1learnsfrominterASprotocolthatsubnetxis
reachablefromAS3andfromAS2.
toconfigureforwardingtable,router1dmustdeterminetowards
whichgatewayitshouldforwardpacketsfordestx.
r thisisalsojobofinterASroutingprotocol!
3c
3b
3a
AS3
1a
x
2a
1c
1d
1b
2c
AS2
2b
AS1
Network Layer
4-
Example:ChoosingamongmultipleASes
nowsupposeAS1learnsfrominterASprotocolthatsubnetxis
reachablefromAS3andfromAS2.
toconfigureforwardingtable,router1dmustdeterminetowards
whichgatewayitshouldforwardpacketsfordestx.
r thisisalsojobofinterASroutingprotocol!
hotpotatorouting:sendpackettowardsclosestoftworouters.
Learn from inter-AS
protocol that subnet
x is reachable via
multiple gateways
Use routing info
from intra-AS
protocol to
determine
costs of least-cost
paths to each
of the gateways
Hot potato routing:
Choose the
gateway
that has the
smallest least cost
Determine from
forwarding table the
interface I that leads
to least-cost gateway.
Enter (x,I) in
forwarding table
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
IntraASRouting
alsoknownasInteriorGatewayProtocols(IGP)
mostcommonIntraASroutingprotocols:
r
RIP:RoutingInformationProtocol
r
OSPF:OpenShortestPathFirst
r
IGRP:InteriorGatewayRoutingProtocol(Ciscoproprietary)
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
RIP(RoutingInformationProtocol)
distancevectoralgorithm
includedinBSDUNIXDistributionin1982
distancemetric:#ofhops(max=15hops)
FromrouterAtosubnets:
u
v
A
z
C
B
D
w
x
y
destinationhops
u1
v2
w2
x3
y3
z2
Network Layer
4-
RIPadvertisements
distancevectors:exchangedamongneighborsevery30sec
viaResponseMessage(alsocalledadvertisement)
eachadvertisement:listofupto25destinationsubnets
withinAS
Network Layer
4-
RIP:Example
z
w
A
x
D
B
y
C
Destination Network
Next Router
Num. of hops to dest.
w
y
z
x
A
B
B
--
2
2
7
1
.
.
....
Routing/ForwardingtableinD
Network Layer
4-
RIP:Example
Dest
w
x
z
.
Next hops
-1
-1
C4
...
w
A
Advertisement
from A to D
z
x
Destination Network
D
B
C
Next Router
y
Num. of hops to dest.
w
y
z
x
A
B
BA
--
2
2
75
1
.
.
....
R o u ting /Fo rwa rd in g ta b le in D
Network Layer
4-
RIP:LinkFailureandRecovery
Ifnoadvertisementheardafter180sec>neighbor/linkdeclareddead
r routesvianeighborinvalidated
r newadvertisementssenttoneighbors
r neighborsinturnsendoutnewadvertisements(iftableschanged)
r linkfailureinfoquickly(?)propagatestoentirenet
r poisonreverseusedtopreventpingpongloops(infinitedistance=
16hops)
Network Layer
4-
RIPTableprocessing
RIProutingtablesmanagedbyapplication-levelprocess
calledrouted(daemon)
advertisementssentinUDPpackets,periodicallyrepeated
routed
routed
Transprt
(UDP)
networkforwarding
(IP)table
Transprt
(UDP)
forwarding
table
network
(IP)
link
physical
link
physical
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
OSPF(OpenShortestPathFirst)
open:publiclyavailable
usesLinkStatealgorithm
r LSpacketdissemination
r topologymapateachnode
r routecomputationusingDijkstrasalgorithm
OSPFadvertisementcarriesoneentryperneighborrouter
advertisementsdisseminatedtoentireAS(viaflooding)
r carriedinOSPFmessagesdirectlyoverIP(ratherthanTCPorUDP
Network Layer
4-
OSPFadvancedfeatures(notinRIP)
security:allOSPFmessagesauthenticated(topreventmalicious
intrusion)
multiplesamecostpathsallowed(onlyonepathinRIP)
Foreachlink,multiplecostmetricsfordifferentTOS(e.g.,satellitelink
costsetlowforbesteffort;highforrealtime)
integrateduniandmulticastsupport:
r MulticastOSPF(MOSPF)usessametopologydatabaseasOSPF
hierarchicalOSPFinlargedomains.
Network Layer
4-
HierarchicalOSPF
Network Layer
4-
HierarchicalOSPF
twolevelhierarchy:localarea,backbone.
Linkstateadvertisementsonlyinarea
r eachnodeshasdetailedareatopology;onlyknowdirection(shortest
path)tonetsinotherareas.
areaborderrouters: summarizedistancestonetsinownarea,
advertisetootherAreaBorderrouters.
backbonerouters:runOSPFroutinglimitedtobackbone.
boundaryrouters:connecttootherASs.
r
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
InternetinterASrouting:BGP
BGP(BorderGatewayProtocol):thedefactostandard
BGPprovideseachASameansto:
1.
2.
3.
ObtainsubnetreachabilityinformationfromneighboringASs.
PropagatereachabilityinformationtoallASinternalrouters.
Determinegoodroutestosubnetsbasedonreachability
informationandpolicy.
allowssubnettoadvertiseitsexistencetorestof
Internet:Iamhere
Network Layer
4-
BGPbasics
pairsofrouters(BGPpeers)exchangeroutinginfooversemi
permanentTCPconnections:BGPsessions
r BGPsessionsneednotcorrespondtophysicallinks.
whenAS2advertisesaprefixtoAS1:
r AS2promisesitwillforwarddatagramstowardsthatprefix.
r AS2canaggregateprefixesinitsadvertisement
3c
3b
eBGPsession
3a
AS3
1a
AS1
iBGPsession
2a
1c
1d
1b
2c
AS2
2b
Network Layer
4-
Distributingreachabilityinfo
usingeBGPsessionbetween3aand1c,AS3sendsprefix
reachabilityinfotoAS1.
r 1ccanthenuseiBGPdodistributenewprefixinfotoall
routersinAS1
r 1bcanthenreadvertisenewreachabilityinfotoAS2over1b
to2aeBGPsession
whenrouterlearnsofnewprefix,itcreatesentryforprefixinits
forwardingtable.
3c
3b
eBGPsession
3a
AS3
1a
AS1
iBGPsession
2a
1c
1d
1b
2c
AS2
2b
Network Layer
4-
Pathattributes&BGProutes
advertisedprefixincludesBGPattributes.
r prefix+attributes=route
twoimportantattributes:
ASPATH:containsASsthroughwhichprefixadvertisementhas
passed:e.g,AS67,AS17
r NEXTHOP:indicatesspecificinternalASroutertonexthopAS.
(maybemultiplelinksfromcurrentAStonexthopAS)
r
whengatewayrouterreceivesrouteadvertisement,uses
importpolicytoaccept/decline.
Network Layer
4-
BGProuteselection
routermaylearnaboutmorethan1routetosomeprefix.
Routermustselectroute.
eliminationrules:
1.
2.
3.
4.
localpreferencevalueattribute:policydecision
shortestASPATH
closestNEXTHOProuter:hotpotatorouting
additionalcriteria
Network Layer
4-
BGPmessages
BGPmessagesexchangedusingTCP.
BGPmessages:
OPEN:opensTCPconnectiontopeerandauthenticatessender
r UPDATE:advertisesnewpath(orwithdrawsold)
r KEEPALIVEkeepsconnectionaliveinabsenceofUPDATES;
alsoACKsOPENrequest
r NOTIFICATION:reportserrorsinpreviousmsg;alsousedtoclose
connection
r
Network Layer
4-
BGProutingpolicy
le g e nd :
B
W
X
A
p ro vid e r
ne two
rk
c us to m e r
ne two rk:
C
Y
A,B,Careprovidernetworks
X,W,Yarecustomer(ofprovidernetworks)
Xisdualhomed:attachedtotwonetworks
XdoesnotwanttoroutefromBviaXtoC
r ..soXwillnotadvertisetoBaroutetoC
r
Network Layer
4-
BGProutingpolicy(2)
le g e nd :
B
W
X
A
p ro vid e r
ne two
rk
c us to m e r
ne two rk:
C
Y
AadvertisespathAWtoB
BadvertisespathBAWtoX
ShouldBadvertisepathBAWtoC?
Noway!BgetsnorevenueforroutingCBAWsinceneitherW
norCareBscustomers
r BwantstoforceCtoroutetowviaA
r Bwantstorouteonlyto/fromitscustomers!
Network Layer
4r
WhydifferentIntraandInterASrouting?
Policy:
InterAS:adminwantscontroloverhowitstrafficrouted,whoroutes
throughitsnet.
IntraAS:singleadmin,sonopolicydecisionsneeded
Scale:
hierarchicalroutingsavestablesize,reducedupdatetraffic
Performance:
IntraAS:canfocusonperformance
InterAS:policymaydominateoverperformance
Network Layer
4-
Chapter4:NetworkLayer
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-
BroadcastRouting
deliverpacketsfromsourcetoallothernodes
sourceduplicationisinefficient:
duplicate
duplicate
creation/transmission
R1
R1
duplicate
R2
R2
R3
R4
source
duplication
R3
R4
in-network
duplication
sourceduplication:howdoessourcedeterminerecipient
addresses?
Network Layer
4-
Innetworkduplication
flooding:whennodereceivesbrdcstpckt,sendscopytoall
neighbors
r
Problems:cycles&broadcaststorm
controlledflooding:nodeonlybrdcstspktifithasnt
brdcstsamepacketbefore
Nodekeepstrackofpcktidsalreadybrdcsted
r Orreversepathforwarding(RPF):onlyforwardpcktifit
arrivedonshortestpathbetweennodeandsource
r
spanningtree
r Noredundantpacketsreceivedbyanynode
Network Layer
4-
SpanningTree
Firstconstructaspanningtree
Nodesforwardcopiesonlyalongspanningtree
A
B
c
F
A
E
B
c
D
F
E
D
G
(a) Broadcast initiated at A
G
(b) Broadcast initiated at D
Network Layer
4-
SpanningTree:Creation
Centernode
Eachnodesendsunicastjoinmessagetocenternode
r Messageforwardeduntilitarrivesatanodealreadybelongingtospanning
tree
A
A
3
B
c
4
F
1
2
E
B
c
D
F
5
E
D
G
G
(a) Stepwise construction
of spanning tree
(b) Constructed spanning
tree
Network Layer
4-
MulticastRouting:ProblemStatement
Goal:findatree(ortrees)connectingroutershavinglocal
mcastgroupmembers
r
r
r
tree:notallpathsbetweenroutersused
sourcebased:differenttreefromeachsendertorcvrs
sharedtree:sametreeusedbyallgroupmembers
Sharedtree
Sourcebasedtrees
Approachesforbuildingmcasttrees
Approaches:
sourcebasedtree:onetreepersource
shortestpathtrees
r reversepathforwarding
r
groupsharedtree:groupusesonetree
r minimalspanning(Steiner)
r centerbasedtrees
wefirstlookatbasicapproaches,thenspecificprotocolsadoptingthese
approaches
ShortestPathTree
mcastforwardingtree:treeofshortestpathroutesfrom
sourcetoallreceivers
r
Dijkstrasalgorithm
S:source
R1
1
2
LEGEND
R4
R2
3
R3
routerwithattached
groupmember
5
4
R6
routerwithnoattached
groupmember
R5
6
R7
i
linkusedforforwarding,
iindicatesorderlink
addedbyalgorithm
ReversePathForwarding
relyonroutersknowledgeofunicastshortestpathfromit
tosender
eachrouterhassimpleforwardingbehavior:
if (mcastdatagramreceivedonincominglinkonshortest
pathbacktocenter)
thenflooddatagramontoalloutgoinglinks
elseignoredatagram
ReversePathForwarding:example
S:source
LEGEND
R1
R4
routerwithattached
groupmember
R2
R5
R3
R6
R7
resultisasourcespecificreverseSPT
maybeabadchoicewithasymmetriclinks
routerwithnoattached
groupmember
datagramwillbeforwarded
datagramwillnotbe
forwarded
ReversePathForwarding:pruning
forwardingtreecontainssubtreeswithnomcastgroupmembers
noneedtoforwarddatagramsdownsubtree
r prunemsgssentupstreambyrouterwithnodownstreamgroup
members
r
LEGEND
S:source
R1
routerwithattached
groupmember
R4
R2
P
R5
R3
R6
P
R7
P
routerwithnoattached
groupmember
prunemessage
linkswithmulticast
forwarding
SharedTree:SteinerTree
SteinerTree:minimumcosttreeconnectingallrouters
withattachedgroupmembers
problemisNPcomplete
excellentheuristicsexists
notusedinpractice:
computationalcomplexity
r informationaboutentirenetworkneeded
r monolithic:rerunwheneverarouterneedstojoin/leave
r
Centerbasedtrees
singledeliverytreesharedbyall
onerouteridentifiedascenteroftree
tojoin:
r edgeroutersendsunicastjoinmsgaddressedtocenterrouter
r joinmsgprocessedbyintermediateroutersandforwarded
towardscenter
r joinmsgeitherhitsexistingtreebranchforthiscenter,or
arrivesatcenter
r pathtakenbyjoinmsgbecomesnewbranchoftreeforthis
router
Centerbasedtrees:anexample
SupposeR6chosenascenter:
LEGEND
R1
3
R2
routerwithattached
groupmember
R4
2
R5
R3
1
R6
R7
1
routerwithnoattached
groupmember
pathorderinwhichjoinmessages
generated
InternetMulticastingRouting:DVMRP
DVMRP:distancevectormulticastroutingprotocol,
RFC1075
floodandprune:reversepathforwarding,sourcebased
tree
RPFtreebasedonDVMRPsownroutingtablesconstructed
bycommunicatingDVMRProuters
r noassumptionsaboutunderlyingunicast
r initialdatagramtomcastgroupfloodedeverywhereviaRPF
r routersnotwantinggroup:sendupstreamprunemsgs
r
DVMRP:continued
softstate:DVMRProuterperiodically(1min.)forgets
branchesarepruned:
mcastdataagainflowsdownunprunedbranch
r downstreamrouter:repruneorelsecontinuetoreceivedata
r
routerscanquicklyregrafttotree
r followingIGMPjoinatleaf
oddsandends
r commonlyimplementedincommercialrouters
r MboneroutingdoneusingDVMRP
Tunneling
Q:Howtoconnectislandsofmulticastroutersinasea
ofunicastrouters?
physicaltopology
logicaltopology
mcastdatagramencapsulatedinsidenormal(nonmulticastaddressed)datagram
normalIPdatagramsentthrutunnelviaregularIPunicasttoreceivingmcast
router
receivingmcastrouterunencapsulatestogetmcastdatagram
PIM:ProtocolIndependentMulticast
notdependentonanyspecificunderlyingunicastroutingalgorithm
(workswithall)
twodifferentmulticastdistributionscenarios:
Dense:
groupmembersdensely
Sparse:
#networkswithgroupmembers
packed,incloseproximity.
smallwrt#interconnectednetworks
bandwidthmoreplentiful
groupmemberswidelydispersed
bandwidthnotplentiful
ConsequencesofSparseDenseDichotomy:
Dense
groupmembershipbyrouters
Sparse:
nomembershipuntilrouters
assumeduntilroutersexplicitly
prune
datadrivenconstructionon
mcasttree(e.g.,RPF)
bandwidthandnongrouprouter
processingprofligate
explicitlyjoin
receiverdrivenconstructionof
mcasttree(e.g.,centerbased)
bandwidthandnongrouprouter
processingconservative
PIMDenseMode
floodandpruneRPF,similartoDVMRPbut
underlyingunicastprotocolprovidesRPFinfoforincoming
datagram
lesscomplicated(lessefficient)downstreamfloodthan
DVMRPreducesrelianceonunderlyingroutingalgorithm
hasprotocolmechanismforroutertodetectitisaleafnode
router
PIMSparseMode
centerbasedapproach
routersendsjoinmsgto
rendezvouspoint(RP)
r
R1
intermediateroutersupdate
stateandforwardjoin
R2
afterjoiningviaRP,router
canswitchtosourcespecific
tree
r
increasedperformance:less
concentration,shorterpaths
R4
join
R3
join
R5
join
R6
alldatamulticast
fromrendezvous
point
R7
rendezvous
point
PIMSparseMode
sender(s):
unicastdatatoRP,which
distributesdownRProoted
tree
RPcanextendmcasttree
upstreamtosource
RPcansendstopmsgifno
attachedreceivers
r
nooneislistening!
R1
R4
join
R2
R3
join
R5
join
R6
alldatamulticast
fromrendezvous
point
R7
rendezvous
point
Chapter4:summary
4.1Introduction
4.2Virtualcircuitand
datagramnetworks
4.3Whatsinsidearouter
4.4IP:InternetProtocol
r
r
r
r
Datagramformat
IPv4addressing
ICMP
IPv6
4.5Routingalgorithms
r Linkstate
r DistanceVector
r Hierarchicalrouting
4.6RoutingintheInternet
r RIP
r OSPF
r BGP
4.7Broadcastandmulticast
routing
Network Layer
4-