Download ch4_NETW_0708_minerva

Communication Networks
P. Demeester
Chapter 4
Network Layer
Computer networking A top-down approach featuring the
internet
4th Edition, 2008
Addison Wesley
James F. Kurose, Keith W. Ross
ISBN 0-321-49770-8
Network Layer
Part of slides provided by J.F Kurose and K.W. Ross, All Rights Reserved
4-1
Chapter 4 outline
4.1 Introduction and Network Service Models
4.2 Routing Principles
4.3 Hierarchical Routing
4.4 The Internet (IP) Protocol
4.5 Routing in the Internet
4.6 What’s Inside a Router
4.7 IPv6
4.8 Multicast Routing
4.9 Mobility
Network Layer
4-2
Network layer functions
 transport packet from
sending to receiving hosts
 network layer protocols in
every host, router
three important functions:
 path determination: route
taken by packets from
source to dest. Routing
algorithms
 forwarding: move packets
from router’s input to
appropriate router output
 call setup: some network
architectures require
router call setup along path
before data flows
application
transport
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
network
data link
physical
application
transport
network
data link
physical
Network Layer
4-3
Network layer functions : IP forwarding
C
Router
Link
W
A
Y:to Y
Y
Routing
Table
Y:to B
Z:to D
Buffer
B
Y:to C
Z:to E
Store
and
Forward
E
X
computer
host
terminal
Source : X
Destination : Y
Content : TCP
D
Z
Network Layer
4-4
Datagram networks: the Internet model
 no call setup at network layer
 routers: no state about end-to-end connections
 no network-level concept of “connection”
 packets forwarded using destination host address
 packets between same source-dest pair may take
different paths
application
transport
network
data link 1. Send data
physical
application
transport
network
2. Receive data
data link
physical
Network Layer
4-5
The IP Service
 End-to-end
 Packet delivery service
 Unreliable, but an honest attempt
 Best-effort, no Quality of Service (QoS)
(no guarantee on delay or bandwidth)
 Designed for internetworking (over any L2)
 Connectionless
Network Layer
4-6
Chapter 4 outline
4.1 Introduction and Network Service Models
4.2 Routing Principles
4.3 Hierarchical Routing
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to destination
 4.4.3 Datagram format
 4.4.4 IP fragmentation
 4.4.5 ICMP: Internet Control Message Protocol
 4.4.6 DHCP: Dynamic Host Configuration Protocol
 4.4.7 NAT: Network Address Translation
4.5 Routing in the Internet
4.6 What’s Inside a Router
4.7 IPv6
4.8 Multicast Routing
4.9 Mobility
Network Layer
4-7
Chapter 4 outline
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to
destination
 4.4.3 Datagram format
 4.4.4 IP fragmentation
 4.4.5 ICMP: Internet Control Message
Protocol
 4.4.6 DHCP: Dynamic Host Configuration
Protocol
 4.4.7 NAT: Network Address Translation
Network Layer
4-8
The Internet Network layer
Host, router network layer functions:
Transport layer: TCP, UDP
Network
layer
IP protocol
•addressing conventions
•datagram format
•packet handling conventions
Routing protocols
•path selection
•RIP, OSPF, BGP
forwarding
table
ICMP protocol
•error reporting
•router “signaling”
Link layer
physical layer
Network Layer
4-9
Address Format - Address Classes
IP address format : 4 octets, decimal notation, separation by dot
example : 157.193.122.10
range : 0 - 255 (corresponds to 0000 0000 - 1111 1111 or 00-FF)
Two parts : network part and host part
157.193.122.10
network
UGent network : 157.193
(64k hosts)
host
[0.0.0.0-128.0.0.0[ [128.0.0.0-192.0.0.0[
HP : 15.0.0.0
UGent : 157.193.0.0
Class A
0 Network
Class B
10
Class C
110
Class D
1110
Class E
11110
Host
Network
128 networks
16 million addr/network
Host
Network
Multicast address
Reserved for future use
address space
Host
16k networks
64k addr/network
2 M networks, 256 addr/network
Network Layer 4-10
Special Addresses
X.Y.0.0 : notation for a network
examples : 15.0.0.0 (class A network), 157.193.0.0 (class B network),
193.125.97.0 (class C network)
127.X.Y.Z : loopback Interface (for debugging)
in practice mainly 127.0.0.1 address used
0.0.0.0 : this host on this network (used for booting)
only allowed as source address
0.0.X.Y : a host on this network (used for booting)
only allowed as source address
255.255.255.255 : broadcast to all hosts on this network
only allowed as destination address, no forwarding allowed
X.Y.255.255 : broadcast to all hosts on a remote network
only allowed as destination address
10.0.0.0-10.255.255.255; 172.16.0.0-172.31.255.255; 192.168.0.0192.168.255.255 :
used for networks not connected to The Internet (e.g. private internets)
A network, subnetwork or host can
NEVER consist entirely of 0’s or 1’s
Network Layer
4-11
Example : UGent network
PC room
Plateau
ftwe01
157.193.103.1
ftwe02
157.193.103.2
ftwe51
157.193.103.51
157.193.103.254
Router Plateau
(ftwe)
Belnet backbone
157.193.227.2Router Belnet
157.193.227.1
157.193.234.2
157.193.60.31
157.193.234.1
(belnet)
Router ARC
(rtr-arc)
157.193.40.254
Router Technicum
157.193.60.254 (rtr-tech)
eduserv1
eduserv2
157.193.40.9 157.193.40.10
Network Layer 4-12
Direct connected networks
PC room
Plateau
ftwe01
157.193.103.1
ftwe02
157.193.103.2
ftwe51
157.193.103.51
direct connected
network
157.193.103.254
Router Plateau
(ftwe)
157.193.234.2
direct
157.193.60.31
direct
connected
connected 157.193.234.1
network
network
Router Technicum
157.193.60.254 (rtr-tech)
Belnet backbone
direct
157.193.227.2Router Belnet
connecte (belnet)
157.193.227.1
d network
Router ARC
(rtr-arc)
direct
157.193.40.254
connected
network
eduserv1
eduserv2
157.193.40.9 157.193.40.10
Network Layer 4-13
Subnetworks
PC room
Plateau
ftwe01
157.193.103.1
ftwe02
157.193.103.2
Subnetwork
157.193.103.0
network
host
157.193.103.254
subnetwork
Router Plateau
(ftwe)
157.193.227.2
Router
Subnetwork
157.193.227.1
157.193.227.0
157.193.234.2
157.193.60.31
Subnetwork
157.193.60.0
157.193.60.254
Belnet backbone
ftwe51
157.193.103.51
Subnetwork
157.193.234.0
157.193.234.1
Belnet
(belnet)
Router ARC
(rtr-arc)
Subnetwork
157.193.40.254
157.193.40.0
Router Technicum
(rtr-tech)
eduserv1
eduserv2
157.193.40.9 157.193.40.10
Network Layer 4-14
Subnet addressing in IP
Without subnetting
Network Host
With subnetting
Network
Subnet mask
1 1 1 1 1 …1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0…0 0 0 0
Subnetwork : 157.193.103.0 (mask FF.FF.FF.00)
network part : 157.193 (class B)
subnetwork part : 103 (8 bits, 254 subnets)
mask used : FF.FF.FF.0000 0000
# hosts : 254 (00 and FF not allowed)
host range : 157.193.103.1 - 157.193.103.254
Subnet
Host
BORDER SUBNET | HOST
Subnetwork : 158.78.42.64 (mask FF.FF.FF.C0)
network part : 158.78 (class B)
subnetwork part: 00101010.01 (10 bits, 1022 subnets)
mask used : FF.FF.FF.1100 0000
# hosts : 62 (00 0000 and 11 1111 not allowed)
host range : 158.78.42.65 - 158.78.42.126
Subnetwork : 158.78.01.00 (mask FF.FF.FF.C0)
host range : 158.78.01.01 - 158.78.01.62
Note : binary,
decimal and
hexadecimal
Network
Layer
4-15
notation used where appropriate
Subnet addressing in IP : class B
Subnetwork : 157.193.0.0 (mask FF.FF.FF.00) : NOT ALLOWED
subnetwork = all 0’s not allowed because no distinction with
the class B address 157.193.0.0 (in practice : used)
Subnetwork : 157.193.255.0 (mask FF.FF.FF.00) : NOT ALLOWED
subnetwork = all 1’s not allowed
(in practice : used)
class B subnetworks
SUBNET
BITS
0
# OF
HOST BITS
SUBNETS
0
16
# OF
HOSTS
65534
255.255.0.0
MASK
1
-
15
-
not allowed
2
2
14
16382
255.255.192.0
3
6
13
8190
4
14
12
4096
255.255.240.0
5
30
11
2046
255.255.248.0
…
…
…
…
…
11
2046
5
30
12
4094
4
14
255.255.255.240
13
8190
3
6
255.255.255.248
14
16382
2
2
255.255.255.252
15
-
1
-
not allowed
255.255.224.0
255.255.255.224
Network Layer 4-16
Subnet addressing in IP : class C
subnetmask (CIDR)
subnetmask (DEC)
subnetmask (HEX)
/24
/25
/26
/27
/28
/29
/30
0
.128
.192
.224
.240
.248
.252
.00
.80
.C0
.E0
.F0
.F8
.FC
0 255 0 127 0 63 0 31 0 15 0 7 0 3
4 7
8 15 8 11
12 15
16 31 16 23 16 19
20 23
24 31 24 27
28 31
32 63 32 47 32 39 32 35
36 39
40 47 40 43
44 47
48 63 48 55 48 51
52 55
56 63 56 59
60 63
64 127 64 95 64 79 64 71 64 67
68 71
72 79 72 75
76 79
80 95 80 87 80 83
84 87
88 95 88 91
92 95
96 127 96 111 96 103 96 99
100 103
104 111 104 107
108 111
112 127 112 119 112 115
116 119
120 127 120 123
124 127
128 255 128 191 128 159 128 143 128 135 128 131
132 135
136 143 136 139
140 143
144 159 144 151 144 147
148 151
152 159 152 155
156 159
160 191 160 175 160 167 160 163
164 167
168 175 168 171
172 175
176 191 176 183 176 179
180 183
184 191 184 187
188 191
192 255 192 223 192 207 192 199 192 195
196 199
200 207 200 203
204 207
208 223 208 215 208 211
212 215
216 223 216 219
220 223
224 255 224 239 224 231 224 227
228 231
232 239 232 235
236 239
240 255 240 247 240 243
244 247
248 255 248 251
Network Layer 252
4-17
255
Subnet addressing class C : example
subnetmask (CIDR)
subnetmask (DEC)
subnetmask (HEX)
/24
/25
/26
/27
/28
/29
/30
0
.128
.192
.224
.240
.248
.252
.00
.80
.C0
.E0
.F0
.F8
.FC
0 255 0 127 0 63 0 31 0 15 0 7 0 3
4 7
8 15 8 11
12 15
16 31 16 23 16 19
20 23
24 31 24 27
28 31
32 63 32 47 32 39 32 35
36 39
40 47 40 43
44 47
48 63 48 55 48 51
52 55
56 63 56 59
60 63
64 127 64 95 64 79 64 71 64 67
68 71
72 79 72 75
76 79
80 95 80 87 80 83
84 87
88 95 88 91
92 95
96 127 96 111 96 103 96 99
100 103
104 111 104 107
108 111
112 127 112 119 112 115
116 119
120 127 120 123
124 127
128 255 128 191 128 159 128 143 128 135 128 131
132 135
136 143 136 139
140 143
144 159 144 151 144 147
148 151
152 159 152 155
156 159
160 191 160 175 160 167 160 163
164 167
168 175 168 171
172 175
176 191 176 183 176 179
180 183
184 191 184 187
188 191
192 255 192 223 192 207 192 199 192 195
196 199
200 207 200 203
204 207
208 223 208 215 208 211
212 215
216 223 216 219
220 223
224 255 224 239 224 231 224 227
228 231
232 239 232 235
236 239
240 255 240 247 240 243
244 247
248 255 248 251
Network Layer 252
4-18
255
Subnet addressing in IP : Example
IF
(destination_address AND subnet_mask)==(my_address AND subnet_mask)
THEN
destination host is on a directly connected network
ELSE
the packet must be sent to a router first
Packet to 157.193.102.51
AND with FF.FF.FF.00
gives : 157.193.102.0
==
157.193.102.3
AND with FF.FF.FF.00
gives : 157.193.102.0
==> LOCAL HOST
157.193.103.12
Subnetwork
157.193.103.0
157.193.103.254
Subnetwork
157.193.102.0
router
157.193.102.254
157.193.102.51
subnet mask : FF.FF.FF.00
157.193.102.3
note : ARP
Packet to 157.193.103.12
AND with FF.FF.FF.00
gives : 157.193.103.0
NOT ==
157.193.102.3
AND with FF.FF.FF.00
gives : 157.193.102.0
==> GO TO ROUTER
Network Layer 4-19
IP addressing: CIDR
 Classful addressing:


inefficient use of address space, address space exhaustion
e.g., class B net allocated enough addresses for 65K hosts,
even if only 2K hosts in that network
 CIDR: Classless InterDomain Routing


network portion of address of arbitrary length
address format: a.b.c.d/x, where x is # bits in network
portion of address
network
part
host
part
11001000 00010111 00010000 00000000
200.23.16.0/23
Network Layer 4-20
IP addresses: how to get one?
Q: How does host get IP address?
 hard-coded by system admin in a file
Wintel: control-panel->network->configuration>tcp/ip->properties
 UNIX: /etc/rc.config
 DHCP: Dynamic Host Configuration Protocol:
dynamically get address from as server
 “plug-and-play”
(more shortly)

Network Layer 4-21
IP addresses: how to get one?
Q: How does network get network part of IP
addr?
A: gets allocated portion of its provider ISP’s
address space
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 4-22
Hierarchical addressing: route aggregation
Hierarchical addressing allows
efficient advertisement of routing information:
Organization 0
200.23.16.0/23
Organization 1
200.23.18.0/23
Organization 2
200.23.20.0/23
Organization 7
.
.
.
.
.
.
ISP - Belnet
“Send me anything
with addresses
beginning
200.23.16.0/20”
Internet
200.23.30.0/23
ISP - Newnet
“Send me anything
with addresses
beginning
199.31.0.0/16”
Network Layer 4-23
Hierarchical addressing: more specific routes
ISPs - Newnet has a more specific route to Organization 1
Organization 0
200.23.16.0/23
Organization 2
200.23.20.0/23
Organization 7
.
.
.
.
.
.
ISP - Belnet
“Send me anything
with addresses
beginning
200.23.16.0/20”
Internet
200.23.30.0/23
ISP - Newnet
Organization 1
200.23.18.0/23
“Send me anything
with addresses
beginning 199.31.0.0/16
or 200.23.18.0/23”
Network Layer 4-24
IP addressing: the last word...
Q: How does an ISP get block of addresses?
A: ICANN: Internet Corporation for Assigned
Names and Numbers
 allocates addresses
 manages DNS
 assigns domain names, resolves disputes
Exc.1
www.icann.org
Network Layer 4-25
Chapter 4 outline
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to
destination : forwarding
 4.4.3 Datagram format
 4.4.4 IP fragmentation
 4.4.5 ICMP: Internet Control Message
Protocol
 4.4.6 DHCP: Dynamic Host Configuration
Protocol
 4.4.7 NAT: Network Address Translation
Network Layer 4-26
Router : Store and Forward
Routing table :
Destination
A
B
Next hop
LAN 2
LAN 3
LAN 2
A
LAN 1
•
•
•
•
A
LAN 3
A
A
B
IP packet arrives in a router (store in a buffer)
Header is analyzed : what is the destination IP address
Look-up of the routing table : next hop (or local)
Forward packet to the correct interface
(or deliver to local application)
Network Layer 4-27
Router Examples
Cisco 12000 series
Network Layer 4-28
Routing Table : example
157.193.103.254
157.193.102.253
157.193.102.254
Subnetwork
157.193.103.0
router B
LAN 1
Subnetwork
157.193.102.0
157.193.102.1
router A
157.193.103.1
157.193.102.34
157.193.104.254
INTERNET
Subnetwork
157.193.104.0
Routing table router A
Destination
Mask
GatewayDestination InterfaceGateway
127.0.0.0
FF.00.00.00
127.0.0.1127.0.0.0/8 Lo0
0.0.0.0
0.0.0.0
157.193.102.253
0.0.0.0/0
LAN 1
157.193.102.253
157.193.102.254
157.193.102.0
FF.FF.FF.00
157.193.102.254
157.193.102.0/24
LAN 1
157.193.102.254
157.193.102.254
157.193.103.0
FF.FF.FF.00
157.193.103.254
157.193.103.0/24
LAN 2
157.193.104.0
FF.FF.FF.00
157.193.104.254
157.193.104.0/24
LAN 3
157.193.103.254
157.193.103.254
Network Layer 4-29
157.193.104.254
157.193.104.254
157.193.104.1
127.0.0.1
Interface
127.0.0.1
Routing Table : Example UGent network
Routing table for an edge router (ftwe)
PC room
Plateau
ftwe01
157.193.103.1
ftwe02
157.193.103.2
lan0
157.193.103.254
Router Plateau
(ftwe)
Destination
127.0.0.1
ftwe51
157.193.103.0
157.193.103.51
157.193.60.0
default
Routing table for a host (eduserv2)
Destination
127.0.0.1
157.193.40.0
default
157.193.234.2
157.193.60.31
157.193.234.1
lan1
Gateway
Interface
127.0.0.1
lo0
Belnet
backbone
157.193.103.254
lan0
157.193.60.31
lan1
157.193.60.254
lan1
Router Technicum
157.193.60.254
(rtr-tech)
Gateway
Interface
157.193.227.2
Router
Belnet
127.0.0.1
lo0
(belnet)
157.193.227.1
157.193.40.10
hme0
157.193.40.254
hme0
Router ARC
(rtr-arc)
157.193.40.254
hme0
eduserv1
eduserv2
157.193.40.9 157.193.40.10
Network Layer 4-30
Routing Table : examples
allserv:/staff/ftwe/pdemeest$
allserv:/staff/ftwe/pdemeest$ netstat
netstat -rn
-r
Routing Table:
Table:
Routing
Destination
Gateway
Destination
Gateway
--------------------------------------- --------------------------------------157.193.40.0
157.193.40.42
157.193.40.0
allserv.rug.ac.be
224.0.0.0
157.193.40.42
BASE-ADDRESS.MCAST.NET
allserv.rug.ac.be
default
157.193.40.254
default
rtr-arc.rug.ac.be
127.0.0.1
127.0.0.1
localhost
localhost
Flags
Use
Interface
Flags Ref
Ref
Use
Interface
--------- --------- ----------- ----------------U
3
U
3 63929
63857 hme0
hme0
U
3
0
U
3
0 hme0
hme0
UG
028820629
UG
028790639
UH
07349737
UH
07347592 lo0
lo0
Flags :
U : route is up
G : route to a gateway (if not set : directly connected destination)
H : route to a host (if not set : route to a network)
C:\WINDOWS>netstat
Route Table
Active Routes:
Network Address
0.0.0.0
127.0.0.0
157.193.122.0
157.193.122.201
157.193.255.255
224.0.0.0
255.255.255.255
-r
Netmask
0.0.0.0
255.0.0.0
255.255.255.0
255.255.255.255
255.255.255.255
224.0.0.0
255.255.255.255
Gateway Address
157.193.122.1
127.0.0.1
157.193.122.201
127.0.0.1
157.193.122.201
157.193.122.201
157.193.122.201
Interface Metric
157.193.122.201
1
127.0.0.1
1
157.193.122.201
1
127.0.0.1
1
157.193.122.201
1
157.193.122.201
1 4-31
Network Layer
157.193.122.201
1
The Forwarding Process
 Sanity checks:
version number and header checksum
 decrement TTL (if reduced to 0 ==> discard)
 Match destination address with:
 own addresses (no forwarding needed)
 prefix(es) in the routing table (longest match)

 Send to outgoing interface:


fragment if packet size > MTU (Maximum Transfer
recalculate header checksum
Unit)
Routing table :
157.193.84.0
direct
lan1
157.193.0.0
157.193.234.2 lan2
destination: 157.193.84.20: longest prefix match is 157.193.84 => lan1
(note : also 157.193 is a prefix match, but not the longest)
Network
destination: 157.193.85.20: longest prefix match is 157.193
=> Layer
lan2 4-32
The Routing Table in a Large Network
European Backbone Router :
• entries to directly attached networks (e.g. BELNET)
• entries to every network attached to these directly attached networks
(e.g.: BELNET : UGent, KULeuven, VUB, … backbone)
• entries to every network in US (unless default)
Dutch Backbone
UGent Backbone
RUGNET
157.193.0.0
KULeuven
Backbone
134.58.0.0
UK Backbone
Belgian Backbone
BELNET
193.190.19x.0
R1
Transatlantic
R2
Large Router :
• up to 40.000 entries in routing table
• long search for appropriate entry
• stability of routing protocols
(many routes to be advertised)
European Backbone
R4
R3
German
Backbone
French Backbone
Network Layer
UGent : 1997
Network Layer 4-34
UGent : 2000
.4
.6
.13
.5
subnet :
157.193.188.0
(FF.FF.FF.00)
.9
.8
.10
.7
.2
.12
.1
.3
.11
rtr-urb
.184.0
hera
URBIS
tacitus
.122.0
Network Layer 4-35
UGent : 2003
Network Layer 4-36
BELNET Backbone
Network Layer 4-37
BELNET Backbone
Network Layer 4-38
GEANT Backbone
Exc.2
Network Layer 4-39
Chapter 4 outline
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to
destination

4.4.3 Datagram format
4.4.4 IP fragmentation
 4.4.5 ICMP: Internet Control Message
Protocol
 4.4.6 DHCP: Dynamic Host Configuration
Protocol
 4.4.7 NAT: Network Address Translation

Network Layer 4-40
IP Packet Format
Type of Service
Currently : IP version Number
4 or IPv4of 32 bit words3 bits for priority
(if no options : 5) 4 bits for delay, BW, …
also IPv6
Number of bytes in IP packet
(including header)
maximum of 65535 bytes
typical few 100 to few 1000
1 bit reserved
NOT USED
4-bit
4-bit
8-bitToS
version header
length
16-bit identification
8-bit TTL
16-bit total length of packet
3-bit flags
13-bit fragment
offset
8-bit protocol
16-bit header checksum
32-bit source IP address
32 bit source and destination address
Calculated over header
32-bit destination IP address
similar to TCP
Used for fragmentation
Identification of each packet
send by host
Indicate higher layer protocol
(increment each time packet Not often used (timestamp, route to follow, …)
Options
(if any)
Internet
Control
Message Prot.
is sent)
Time-To-Live :
start at 32 or 64
decrement at each hop
discard packet if TTL=0
1
2
4
6
17
46
89
ICMP
Internet Group Management Prot. IGMP
IP-in-IP tunnelling
IP-in-IP
Transmission
Data Control Protocol TCP
User Datagram Protocol
UDP
Resource Reservation Protocol RSVP
Open Shortest Path First
OSPF
Network Layer 4-41
Chapter 4 outline
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to
destination
 4.4.3 Datagram format

4.4.4 IP fragmentation
4.4.5 ICMP: Internet Control Message
Protocol
 4.4.6 DHCP: Dynamic Host Configuration
Protocol
 4.4.7 NAT: Network Address Translation

Network Layer 4-42
IP Fragmentation & Reassembly
 network links have MTU
(max.transfer size) - largest
possible link-level frame.
 different link types,
different MTUs
 large IP datagram divided
(“fragmented”) within network
 one datagram becomes
several datagrams
 “reassembled” only at final
destination
 IP header bits used to
identify, order related
fragments
fragmentation:
in: one large datagram
out: 3 smaller datagrams
reassembly
Network Layer 4-43
IP Fragmentation and Reassembly
Example
 4000 byte
datagram
 MTU = 1500 bytes
length ID fragflag offset
=4000 =x
=0
=0
One large datagram becomes
several smaller datagrams
Header (20 bytes)
length ID fragflag offset
=1500 =x
=1
=0
length ID fragflag offset
=1500 =x
=1
=1480
length ID fragflag offset
=1040 =x
=0
=2960
Note : - offset should be multiple of 8 (because encoded in 13 bits)
Network Layer
- for offset : do not take header into account
4-44
Chapter 4 outline
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to
destination
 4.4.3 Datagram format
 4.4.4 IP fragmentation
4.4.5 ICMP: Internet Control Message
Protocol

4.4.6 DHCP: Dynamic Host Configuration
Protocol
 4.4.7 NAT: Network Address Translation
Network Layer 4-45

ICMP : Internet Control Message Protocol
ICMP resolves the following questions :
• What happens after a router crashed and disrupted the network
• What notice is given that datagrams are wandering around until
their Time-To-Live (TTL) expires ?
• What warning is given to the application when sending information
to an unreachable host ?
• How to know neighbors (routers) ?
• ...
• How can we know that a host is up and running ?
• ...
IP-datagram
IP-header
ICMP message
protocol =1
ICMP is often considered part of the IP layer
Network Layer 4-46
ICMP Message
0
15 16
8 9
type
code
31
checksum
message content
(depends on type and code)
ERROR :
type = 3 : destination unreachable
code
code
code
code
code
…
= 0 : network unreachable
= 1 : host unreachable
= 2 : protocol unreachable
= 3 : port unreachable
= 4 : fragmentation needed but
don’t-fragment bit set
type = 5 : redirect
code = 0 : redirect for network
code = 1 : redirect for host
…
QUERY :
type = 0, code = 0 : echo reply (ping reply)
type = 8, code = 0 : echo request (ping request)
type = 9, code = 0 : router advertisement
type = 10, code = 0 : router solicitation
...
type = 11 : time exceeded
code = 0 : TTL = 0 during transit
code = 1 : TTL = 0 during reassembly
Network Layer 4-47
ICMP query : Echo (and use in ping)
0
89
type
code
identifier
15 16
31
checksum
sequence number
data
ICMP Ping Request
ICMP Ping Reply
C:\WINDOWS>ping www.kpn.com
C:\WINDOWS>ping
Usage:
ping [-t] [145.7.233.150]
[-a] [-n count] with
[-l size]
[-f]of[-i TTL] [-v TOS]
Pinging
www.kpn.com
32 bytes
[-r count] [-s count] [[-j host-list] | [-k host-list]
[-w timeout] destination-list
Reply
from 145.7.233.150: bytes=32 time=93ms TTL=239
Options:
-t 145.7.233.150:
Pingbytes=32
the specifed
host TTL=239
until interrupted.
Reply from
time=55ms
-a
Resolve addresses to hostnames.
Reply from
145.7.233.150:
bytes=32
time=54ms
-n count
Number
of echo
requestsTTL=239
to send.
-l size
Sendbytes=32
buffer size.
Reply from
145.7.233.150:
time=59ms TTL=239
-f
Set Don't Fragment flag in packet.
-i TTL
Time To ping
Live.-s www.kpn.com
allserv:/staff/ftwe/pdemeest$
-v TOS
Type Of Service.
PING www.kpn.com:
bytes
-r count 56 data
Record
route for count hops.
count
Timestamp
for count hops.
64 bytes-sfrom
www.kpn.com
(145.7.233.150):
icmp_seq=0. time=24. ms
-j
host-list
Loose
source
route
along
host-list.
64 bytes-kfrom
www.kpn.com
(145.7.233.150):
icmp_seq=1.
time=28. ms
host-list
Strict
source route along
host-list.
Network Layer 4-48
64 bytes-wfrom
www.kpn.com
(145.7.233.150):
icmp_seq=2.
time=28.
ms
timeout
Timeout
in milliseconds
to wait for
each reply.
ICMP error : Redirect
INTERNET
157.193.138.2
Routing Table :
default
157.193.138.0
157.193.122.0
157.193.184.0
157.193.138.2
157.193.138.1
157.193.122.1
157.193.122.2
157.193.184.0
157.193.138.1
B
A
.1
.2
2 : forward first packet
157.193.122.0
1 : first packet
.51 4 : next packets
3 : ICMP redirect
Send IP packet to 157.193.184.24
Routing Table :
default
157.193.122.0
157.193.184.24
157.193.122.1
157.193.122.51
157.193.122.2
0
15 16
8 9
type = 5
code
31
checksum
157.193.122.2
address of
router to be used
a.o.: 157.193.184.24
IP header + 8 data octets
Network Layer 4-49
of original datagram
ICMP error : Time Exceeded use in traceroute
GOAL : find-out the route to a destination (and some additional info)
• host running traceroute sends UDP IP-packet to destination with TTL = 1
• first router : decrement TTL to 0, discard packet, ICMP time exceeded back
• hosts sends UDP IP-packet to destination with TTL = 2
• first router : decrement TTL to 1, second router : decrement TTL to 0,
discard packet, ICMP time exceeded back
• ...
allserv:/staff/ftwe/pdemeest$ traceroute www.intec.rug.ac.be
traceroute to intecsrv.rug.ac.be (157.193.92.92), 30 hops max, 40 byte
packets
1 rtr-arc.rug.ac.be (157.193.40.254) 1 ms 1 ms 1 ms
2 rtr-tech.rug.ac.be (157.193.138.10) 1 ms 1 ms 1 ms
3 intecsrv.rug.ac.be (157.193.92.92) 4 ms 1 ms 3 ms
TTL
router name
allserv
IP/UDP TTL=1
router IP-address
rtr-arc
ICMP Time Exceeded
rtr-arc
roundtrip time (3x)
rtr-tech
intecsrv
Network Layer 4-50
Chapter 4 outline
4.4 The Internet (IP) Protocol
 4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to destination
 4.4.3 Datagram format
 4.4.4 IP fragmentation
 4.4.5 ICMP: Internet Control Message Protocol
4.4.6 DHCP: Dynamic Host Configuration
Protocol
 4.4.7 NAT: Network Address Translation

Network Layer 4-51
Home Network : single PC
HOME
ADSL
(Belgacom) /
Cable Modem
(Telenet)
145.122.201.1
IP address : 157.193.0.1
gateway : 157.193.0.254
DNS 1 : 145.122.201.1
Ethernet
public
gateway
router
or dynamic configuration :
DHCP (Dynamic Host Control Protocol)
DNS
INTERNET
157.193.0.254
157.193.0.1
Network Layer 4-52
Home Network : multiple PC’s
HOME
IP address : 192.168.0.2
gateway : 192.168.0.1
DNS 1 : 145.122.201.1
ADSL
(Belgacom) /
Cable Modem
(Telenet)
private
network
192.168.0.0
145.122.201.1
192.168.0.2
switch
private
gateway
router
public
gateway
router
DNS
INTERNET
192.168.0.3
192.168.0.1
157.193.0.254
157.193.0.1
192.168.0.4
Network Layer 4-53
NAT : Network Address Translation
WWW
server
157.193.0.1:2012 =>
202.87.122.4:80
192.168.0.2:1093 =>
202.87.122.4:80
202.87.122.4:80 =>
157.193.0.1:2012
202.87.122.4:80 =>
192.168.0.2:1093
202.87.122.4
Port 2012 == 192.168.0.2:1093
192.168.0.2
Port 2015 == 192.168.0.4:1212
INTERNET
NAT
only known address
in Internet !
192.168.0.1
192.168.0.3
157.193.0.254
157.193.0.1
202.87.122.4:80 =>
192.168.0.4:1212
192.168.0.4 192.168.0.4:1212 =>
202.87.122.4:80
202.87.122.4:80 =>
157.193.0.1:2015
157.193.0.1:2015 =>
Network Layer 4-54
202.87.122.4:80
NAT: Network Address Translation
 Motivation: local network uses just one IP address as
far as outside world is concerned:
 no need to be allocated range of addresses from ISP:
- just one IP address is used for all devices
 can change addresses of devices in local network
without notifying outside world
 can change ISP without changing addresses of
devices in local network
 devices inside local net not explicitly addressable,
visible by outside world (a security plus).
Network Layer 4-55
NAT: Network Address Translation
 16-bit port-number field:

>60,000 simultaneous connections with a single
LAN-side address!
 NAT is controversial:
 routers
should only process up to layer 3
 violates end-to-end argument
• NAT possibility must be taken into account by app
designers, e.g., P2P applications
 address
IPv6
shortage should instead be solved by
Network Layer 4-56
NAT : Network Address Translation
User
PC
No packets may enter unless in table
Example : WWW server on 192.168.0.2:1321
NAT should open port to enter !
202.87.122.4
202.87.122.4:4520 =>
157.193.0.1:80
192.168.0.2
Port 80 == 192.168.0.3:1321
WWW server
at port 1321
192.168.0.3
INTERNET
202.87.122.4:4520 =>
192.168.0.3:1321
NAT
192.168.0.1
157.193.0.254
157.193.0.1
!!! NAT works as firewall !!!
192.168.0.4
Network Layer 4-57
DHCP : Dynamic Host Configuration Protocol
IP : 192.168.0.13
gateway : 192.168.0.1
DNS : 145.122.201.1
192.168.0.13
DHCP request
145.122.201.1
192.168.0.13
DHCP
server
DNS
192.168.0.23
192.168.0.42
INTERNET
192.168.0.23
192.168.0.1
157.193.0.254
157.193.0.1
Easy configuration of PCs
192.168.0.42
Network Layer 4-58
How do the PCs know the DHCP server ? ==>DHCP discover
DHCP : Dynamic Host Configuration Protocol
157.193.0.5
157.193.0.1
157.193.0.5
157.193.0.1
145.122.201.1
DNS
DHCP
server
INTERNET
IP : 157.193.0.1
gateway : 157.193.0.254
DNS : 145.122.201.1
157.193.0.254
reuse of limited number of IP numbers
Network Layer 4-59
DHCP: Dynamic Host Configuration Protocol
Goal: allow host to dynamically obtain its IP address
from network server when it joins network
Can renew its lease on address in use
Allows reuse of addresses (only hold address while connected
and “on”)
Support for mobile users who want to join network
DHCP overview:
 host broadcasts “DHCP discover” msg
 DHCP server responds with “DHCP offer” msg
 host requests IP address: “DHCP request” msg
 DHCP server sends address: “DHCP ack” msg
Network Layer 4-60
DHCP client-server scenario
DHCP server: 223.1.2.5
src : 0.0.0.0, 68
dest.: 255.255.255.255,67
DHCPDISCOVER
yiaddr: 0.0.0.0
transaction ID: 654
arriving
client
src: 223.1.2.5, 67
dest: 255.255.255.255, 68
DHCPOFFER
yiaddrr: 223.1.2.4
transaction ID: 654
DHCP server ID : 233.1.2.5
Lifetime: 3600 secs
time
src: 0.0.0.0, 68
dest:: 255.255.255.255, 67
DHCPREQUEST
yiaddrr: 223.1.2.4
transaction ID: 655
DHCP server ID : 233.1.2.5
Lifetime: 3600 secs
Standard ports : 67, 68
yiaddrr : Your Internet Address
src: 223.1.2.5, 67
dest: 255.255.255.255, 68
DHCPACK
yiaddrr: 223.1.2.4
transaction ID: 655
DHCP server ID : 233.1.2.5
Lifetime: 3600 secs
Network Layer 4-61
Chapter 4 outline
4.1 Introduction and Network Service Models
4.2 Routing Principles
4.3 Hierarchical Routing
4.4 The Internet (IP) Protocol
4.5 Routing in the Internet
 4.5.1 Intra-AS routing: RIP and OSPF
 4.5.2 Inter-AS routing: BGP
4.6 What’s Inside a Router?
4.7 IPv6
4.8 Multicast Routing
4.9 Mobility
Network Layer 4-62
Routing <> Forwarding
Routing : fill in the routing tables of a router,
based on centralized or distributed calculation
of a shortest route or distance between router and destination network
Forwarding : forward the packets,
based on the destination IP address
and the routing table content
A
W
C
Y
Routing
Table ?
B
Y:to B
Destination : Y
shortest
route
Forward
D
X
distance
to Y : 5
distance
to Y : 3
E
Network Layer 4-63
Routing in the Internet (hierarchical)
scale: with 200 million destinations:
 can’t store all dest’s in routing tables!
 routing table exchange would swamp links!
 The Global Internet consists of Autonomous Systems (AS,
aggregation of routers in regions) interconnected with each
other:



Stub AS: small corporation: one connection to other AS’s
Multihomed AS: large corporation (no transit): multiple connections
to other AS’s
Transit AS: provider, hooking many AS’s together
 Two-level routing:
 Intra-AS: administrator responsible for choice of routing algorithm
within network (administrative autonomy), may be different in
different AS’s
Network Layer 4-64
 Inter-AS: unique standard for inter-AS routing: BGP
Intra-AS and Inter-AS routing
Gateways:
C.b
a
C
B.a
A.a
b
A.c
d
A
a
b
inter-AS, intra-AS
routing in
gateway A.c
c
a
c
B
b
•perform inter-AS
routing amongst
themselves
•perform intra-AS
routing with other
routers in their
AS
network layer
data link layer
physical layer
Network Layer 4-65
Intra-AS and Inter-AS routing
C.b
a
Host
h1
C
b
A.a
Inter-AS
routing
between
A and B
A.c
a
d
c
b
A
Intra-AS routing
within AS A
B.a
a
c
B
Host
h2
b
Intra-AS routing
within AS B
Network Layer 4-66
Intra-AS Routing
 Also known as Interior Gateway Protocols (IGP)
 Most common Intra-AS routing protocols:

RIP: Routing Information Protocol

OSPF: Open Shortest Path First

IGRP: Interior Gateway Routing Protocol
(Cisco proprietary)
Network Layer 4-67
RIP ( Routing Information Protocol)
 Distance vector algorithm
 Distributed Bellman-Ford algorithm
 Distance metric: # of hops (max = 15 hops)
 Distance vectors: exchanged among neighbors
every 30 sec via Response Message (also called
advertisement)
 Each advertisement: list of up to 25 destination
nets within AS
 Included in BSD-UNIX Distribution in 1982
Network Layer 4-68
Principle RIP
From B to C: [(A,1);(B,0);(C,1);(D,1);(E,2)]
A
B
E
D
C
destination
A
B
D
E
Next hop
B
direct
direct
D
Interface Distance
BC
2
BC
1
CD
1
CD
2
Network Layer
General Characteristics RIP
 Simple protocol, easy to implement
 Often used, even today
 Slow convergence
 Acceptable for small networks or networks
with a simple topology
 Some routing anomalies that cause very slow
convergence: like counting to infinity (see
next slide)
Network Layer
RIP Problem : Counting to Infinity
R1
R1R
R2L
R2/R1R/1
R3/R1R/2
R2
R1/R2R/5
R1/R2L/1
R1/R2R/3
R3/R2R/1
R3/R2R/1
R2R
R3L
[(R1,2);(R2,1);(R3,0)]
[(R1,3);(R2,0);(R3,1)]
[(R1,4);(R2,1);(R3,0)]
[(R2,0);(R3,1)]
R3
R1/R3L/2
R1/R3L/4
R2/R3L/1
Every 30 sec : advertisement
After 180 sec of no adv.: link down
Due to “counting to infinity” problem :
Maximum distance limited to 16
BUT : current large networks can
have more than 16 hops ! ==> only for small networks
(16 hops = “infinity”)
router/interface/distance
Network Layer
RIP Table processing
 RIP routing tables managed by application-level
process called route-d (daemon)
 advertisements sent in UDP packets, periodically
repeated
routed
routed
Transprt
(UDP)
network
(IP)
link
physical
Transprt
(UDP)
forwarding
table
forwarding
table
network
(IP)
link
physical
Network Layer 4-72
Link-State Routing Protocol : OSPF
 Each router obtains an overview of the network
topology in a link state database (exchange of
link-state (LS) packets)
 From this topology (link state database) each
node calculates the shortest path (Dijkstra) to
all destinations
 From the shortest path the next hop is known to
reach a certain destination
 This next hop is used to build the routing table
OSPF : Open Shortest Path First
 “open”: publicly available
 OSPF advertisement carries one entry per neighbor router
 Advertisements disseminated to entire AS (via flooding)

Network Layer
Carried in OSPF messages directly over IP (rather than TCP or UDP)
Link-State Packets
Compose link-state packets Same source router and same sequence number ==> discard
Same source router and older sequence number ==> discard
Flood the link-state packets
Use sequence numbers
[(BC,1),(BD,1),(AB,1)]
A
B
E
D
C
Network Layer
Link-State Routing Protocol Overview
Router C :
Link-State Database
[AB,AE]
incoming
link state
packets
[AB,BD,BC]
[BD,CD,DE]
[AE,DE]
Link
AB
AE
BD
BC
CD
DE
Cost
1
1
1
1
1
1
knowledge of
network topology
From
A
A
B
B
D
D
Dijkstra :
shortest paths
A
B
C
Router C :
Routing Table
Dest.
E
D
A
B
D
E
Next
hop
B
Direct
Direct
D
Interface
BC
BC
CD
CD
Network Layer
OSPF “advanced” features (not in RIP)
 Security: all OSPF messages authenticated (to




prevent malicious intrusion)
Multiple same-cost paths allowed (only one path in
RIP)
For each link, multiple cost metrics for different
TOS (e.g., satellite link cost set “low” for best effort;
high for real time)
Integrated uni- and multicast support:
 Multicast OSPF (MOSPF) uses same topology data
base as OSPF
Hierarchical OSPF in large domains.
Network Layer 4-76
Hierarchical OSPF
backbone
routers
area
border
routers
(AS) boundary
router
BACKBONE
[(10.25.11.0/24,3);(10.20.15.0/24,3)]
[(10.10.0.0/16,3)]
AREA 3
AREA 1
internal
routers
10.10.6.0
/24
AREA 2
10.10.2.0
/24
10.10.3.0
/24
aggregation possible
Backbone = AREA 0
Between areas : always via backbone (“tree”)
One border router may serve multiple areas
One area may use multiple border routers
10.25.11.0
/24
10.20.15.0
/24
NO aggregation possible
Network Layer 4-77
Hierarchical OSPF
 Two-level hierarchy: local area, backbone.
Link-state advertisements only in area
 each nodes has detailed area topology; only know
direction (shortest path) to nets in other areas.
Internal Routers
Area border routers: “summarize” distances to nets
in own area, advertise to other Area Border routers.
Backbone routers: run OSPF routing limited to
backbone.
Boundary routers: connect to other AS’s.





Network Layer 4-78
Chapter 4 outline
4.1 Introduction and Network Service Models
4.2 Routing Principles
4.3 Hierarchical Routing
4.4 The Internet (IP) Protocol
4.5 Routing in the Internet
 4.5.1 Intra-AS routing: RIP and OSPF
 4.5.2 Inter-AS routing: BGP
4.6 What’s Inside a Router?
4.7 IPv6
4.8 Multicast Routing
4.9 Mobility
Network Layer 4-79
Internet inter-AS routing: BGP
Autonomous System
(has AS Number)
OSPF
BGP
BGP
BGP
BGP
RIP
RIP
OSPF
BGP
OSPF
Stub AS
Multi connected AS (but no transit allowed)
Alternative :
Use of default route
(no BGP required)
AS Belnet = multiple domains
Network Layer 4-80
(e.g. UGent, KULeuven,
VUB, …)
I-BGP and E-BGP
E-BGP
I-BGP
E-BGP
I-BGP
E-BGP
E-BGP
E-BGP
I-BGP : used to distribute
BGP information between
Border gateways from single AS
only edge routers
involved in BGP
Network Layer 4-81
BGP : Example
12
77/8
132.2/16
AS numbers : 1-65535
(65412-65535 : private)
Example : Belnet : 2611
AS 54
E-BGP
I-BGP
12/8
AS 54, 88
133.12/16
157.193/16
201.12.12/24
146.123/16
12/8
133.12/16
201.132.15/24
201.12.12/24
54
AS 88
157.193/16
146.123/16
88
201.132.15/24
Stub AS : in general
no AS number
157.193/16
146.123/16
201.132.15/24
NLRI : Network Layer Reachability Information
Network Layer 4-82
Internet inter-AS routing: BGP
 BGP4 (Border Gateway Protocol): the de facto
standard
 Path Vector protocol:
 similar to Distance Vector protocol
 each Border Gateway broadcasts to neighbors
(peers) entire path (i.e., sequence of AS’s) to
destination (more general : NLRI information)
 BGP announces routes to networks (CIDR enabled!),
not individual hosts
Network Layer 4-83
BGP operation
What does a BGP router do?
 Receiving and filtering route advertisements from
directly attached neighbor(s).
 Sending route advertisements to neighbors
 Route selection.
 To route to destination X, which path (of
several advertised) will be taken?
 Filling routing table complex : based on AS_path,
weight attribute, local preference attribute, …
Network Layer 4-84
BGP operation
Suppose: gateway X send its path to peer gateway W
 W may or may not select path offered by X
 cost, policy (don’t route via competitors AS), loop
prevention reasons.
 If W selects path advertised by X, then:
Path (W,Z) = W, Path (X,Z)
 Note: X can control incoming traffic by controlling its
route advertisements to peers:
 e.g., don’t want to route traffic to Z -> don’t
advertise any routes to Z
Network Layer 4-85
BGP messages
 BGP messages exchanged using TCP.
 BGP messages:
OPEN: opens TCP connection to peer and
authenticates sender
 UPDATE: advertises new path (or withdraws old)
 KEEPALIVE keeps connection alive in absence of
UPDATES; also ACKs OPEN request
 NOTIFICATION: reports errors in previous msg;
also used to close connection

Network Layer 4-86
Policy based routing : example 1
12
54
12/8
133.12/16
201.12.12/24
AS 54
12/8
88
133.12/16
201.12.12/24
AS 88 : Dual homed but no transit
 do not forward info from AS 54
Network Layer 4-87
Policy based routing : example 2
77/8
132.2/16
12
54
88
134
12/8
AS 54
133.12/16
12/8
AS 54, 12
201.12.12/24
133.12/16
77/8
AS 54, 12
AS 54
201.12.12/24
132.2/16
77/8
12/8
132.2/16
133.12/16
201.12.12/24
AS 54 transit but only when paid
 do not forward AS 12 AS 134
Network Layer 4-88
BGP in practice: BNIX peering point
Belnet :
- support of research/education networks
- support of BNIX
BNIX : Belgian National Internet eXchange
- place where ISP’s can peer
- connected to >40 ISP’s (see below)
- peering not full mesh (see next)
- other examples : AMS-IX (Nl), SFINX (Fr), LINX (UK), …
Organization
KPN Nederland
Versatel
UUnet
BELNET
AT&T
Arcor AG & Co
Tiscali
PSInet
NETnet
T-Systems
Level3
Easynet
BT Ignite Europe
X-router
Skynet
Perceval
Opentransit
BNIX
194.53.172.71
194.53.172.115
194.53.172.78
194.53.172.65
194.53.172.77
194.53.172.109
194.53.172.97, 194.53.172.105
194.53.172.82
194.53.172.75
194.53.172.113
194.53.172.102
194.53.172.96,194.53.172.98
194.53.172.112
194.53.172.124
194.53.172.81
194.53.172.80
194.53.172.99
MBNIX
BNIX6
194.53.172.254
3FFE:80B0:1000:0:2E0:F7FF:FE25:582B
3FFE:80B0:1000:0:230:96FF:FEE6:DC00
3FFE:80B0:1000:0:204:DDFF:FE08:F10
Network Layer 4-89
3FFE:80B0:1000:0:2D0:BBFF:FE28:7800
BGP : example of peering table BNIX
Not a full mesh !
4: IPv4 peering
6: IPv6 peering
M: Multicast peering
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
ISP
AS
BELNET
2611
UPC/chello
6830
Belgacom
6774
Wanadoo
8277
Easynet
4589
RealROOT
28747
ProServe B.V.
21155
Perceval
5463
Cybernet
13226
XS4ALL Belgium
8201
Openweb
20650
AT&T
2686
Mobistar
12493
Planet Technologies
8737
Brutele
12392
Skynet
5432
LambdaNet
13237
EDPNet
9031
C&W Belgium
6696
Colt
8220
Tiscali
3257
HostIt
9166
M/6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
M6
M6 M6 M6 4
6 4 4 4 4 4 4 4 4 4 4 4M
4 6 4 4
M6 M6
M6 M6 6 M6 6 4 4 4 4 4 4 4 4 4 4 4 M
4M
4
M6 M6 M6
M 6 4 4 4 4 4 4 4M
4 4
4 4 4 4 4 4 4
M6 M6 M6 M
6 4 4 4 4 4 4 4
4 4 4 4 4 4 4
4 4
6 4 6 6 6
4 6 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
M6
M6
4 4 4
4 4 4 4 4 4 4 4 4 4
4 4
4 4 4
6 6 6 4 4 6 4
4 4 4 4 4 4 4 4 4
4 4 4 6 4 4
4 4 4 4 4 4 4
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
4 4 4 4 4 4 4 4
4 4 4 4 4 4 4 4 4
4
4 4
4 4 4 4 4 4 4 4 4
4 4 4 4 4 4 4 4
4 4 4
4 4 4 4 4 4 4 4 4 4
4 4 4 4 4
4 4 4
4 4
4 4 4 4 4 4 4 4 4 4 4
4 4 4 4 4 4 4
4
4
M
4 4M
4 4 4 4 4 4 4 4
4 4 4 4 4 4 4
4 4
4 4 4 4 4 4 4 4 4 4 4 4 4
4 4 4 4
4
4
M
4 4 4 4 4 4 4 4 4 4 4 4 4 4
4 4 4 4 4
4 4
4 4
4 4 4 4 4 4 4 4 4 4 4 4
4 4 4 4 4
4
4 4 4 4 4
4 4 4
4 4 4 4 4
4
4 4 4
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
4 4
4
M M M
4 4 4 4 4 4
4 4 4
4 4
4
4 4
4 4 4 4 4
4 4 4
4
4 4 4 4 4 4 4
4
6 6M
4
4 4 6 4
4
4
4 4
4 4 4-90
Network
Layer
4 4 4 4 4 4 4 4 4 4 4
4 4 4
4 4
4
BGP : traffic in BNIX
> 100% traffic increase
in one year
Network Layer 4-91
Active BGP entries
BGP : active BGP entries
100.000
90
Date
00
(information from Australian
BGP router : bgp.potaroo.net)
Trends :
- increase : smaller domains used (more AS’s), more meshed, …
Network Layer
- improved by CIDR (introduced in 1994-1995)
4-92
AS’s used
BGP : number of AS’s
10.000
97
00
Date
03
If this continues, in a few years time no more AS numbers available
 switch to 32 bit addresses for AS’s (now 16 bit) ?Network Layer 4-93
total address space advertised
BGP : total address space advertised
109
00
Date
03
Network Layer 4-94
total address space advertised (%)
BGP : total address space advertised (%)
25%
00
Date
About 30% of all IPv4 addresses advertised
03
Network Layer 4-95
Why different Intra- and Inter-AS routing ?
Policy:
 Inter-AS: admin wants control over how its traffic is
routed, who routes through its net.
 Intra-AS: single admin, so no policy decisions needed
Scale:
 hierarchical routing saves table size, reduced update
traffic
Performance:
 Intra-AS: can focus on performance
 Inter-AS: policy may dominate over performance
Network Layer 4-96
Chapter 4 outline
4.1 Introduction and Network Service Models
4.2 Routing Principles
4.3 Hierarchical Routing
4.4 The Internet (IP) Protocol
4.5 Routing in the Internet
4.6 What’s Inside a Router
4.7 IPv6
4.8 Multicast Routing
4.9 Mobility
Network Layer 4-97
Table of contents
 4.1 Introduction and Network Service Models
 4.4 The Internet (IP) Protocol
4.4.1 IPv4 addressing
 4.4.2 Moving a datagram from source to destination
 4.4.3 Datagram format
 4.4.4 IP fragmentation
 4.4.5 ICMP: Internet Control Message Protocol
 4.4.6 DHCP: Dynamic Host Configuration Protocol
 4.4.7 NAT: Network Address Translation
 4.5 Routing in the Internet
 4.5.1 Intra-AS routing: RIP and OSPF
 4.5.2 Inter-AS routing: BGP
 Outline
 Table of contents

2
7
8
26
40
42
45
51
51
62
62
79
97
98
Network Layer 4-98