Chapter III

Network Layer
2 Contents

 Introduction
 Internet protocol and IP addressing
 Address mapping, error reporting and Group
Management Protocols
 Routing algorithms
 Asynchronous Transfer Mode (ATM) (Example
for WAN Protocol)
3
4
 Introduction
5

 It is the lowest layer that deals with end-to-end packet (i.e.

Network layer data unit) transmission.
 The first duty of the network layer is packetizing:
encapsulating the payload in a network-layer packet at the
source and decapsulating the payload from the network-layer
packet at the destination.
 Forwarding:- involves the transfer of a packet from an
incoming link to an outgoing link within a single router.
 Routing:- involves all of a network’s routers, whose collective
interactions via routing protocols (algorithms) determine the
paths that packets take on their trips from source to
destination node.
6 Introduction cont….
 Connection setup:
 Connection-oriented communication (virtual circuit
network) and
 Connectionless communication (datagram network)
 Store-and-Forward Packet Switching: A host with a
packet to send transmits it to the nearest router. The
packet is stored there until it has fully arrived and the
link has finished its processing by verifying the checksum.
 Then forwarded to the next router along the path until it
reaches the destination host, where it is delivered. This
mechanism is store-and-forward packet switching.
 Security: network layer provide a security for a
connectionless service, for this purpose we need to have
another virtual level (IPsec) that changes the
connectionless service to a connection-oriented service.
7 Introduction cont….
 Services Provided to the Transport Layer:
 The services provided to the transport layer by the network
layer can be carefully designed by considering the following
transport layer requirements in mind:
[Link] services should be independent of the router technology.
2. The transport layer should be shielded from the number,
type, and topology of the routers present.
3. The network addresses made available to the transport layer
should use a uniform numbering plan, even across LANs and
WANs.
 The network layer should provide connection oriented or
connectionless service to transport layer.
 Connectionless and Connection-oriented
8
Communication
 Comparison of Datagram and Virtual-Circuit
9 Networks
Issue Datagram network Virtual-circuit network
C i r c u i t Not needed Required
setup
Addressin Each packet contains the Each packet contains a short VC
g full source and destination number
address
S t a t e Routers do not hold state Each VC requires router table
informatio i n f o r m a t i o n a b o u t space per connection
n connections
Routing E a c h p a c k e t i s r o u t e d Route chosen when VC is set up;
independently all packets follow it.
E f f e c t o f None, except for packets All VCs that passed through the
r o u t e r lost during the crash failed router are terminated
failures
Quality of Difficult Easy if enough resources can be
service allocated in advance for each VC
Congestion Difficult Easy if enough resources can be
10 Network-Layer Performance Measures
 The performance of a network can be measured in terms
of delay , throughput , and packet loss ( sometimes
congestion control can also be consider to improve the
performance).
 The delays in a network can be divided into four types:
transmission delay, propagation delay, processing delay,
and queuing delay.
 Throughput is the number of bits passing through the
point in a second, which is actually the transmission rate
of data at that point.
 Another issue that severely affects the performance of
communication is the number of packets lost during
transmission.
 Introduction cont….
11
Router
 Router cont….
12
 Two main functions:
 Run routing algorithms/protocols (e.g., RIP, OSPF,
IGRP and others).
 Forwarding datagrams from incoming to outgoing
links
 Input ports
13  An input port performs the following key functions.
 It performs the physical layer function of terminating an
incoming physical link at a router.
 An input port also performs link-layer functions needed to
interoperate with the link layer at the other side of the
incoming link, and
 The lookup function is also performed at the input port. It is
here that the forwarding table is consulted to determine the
router output port to which an arriving packet will be
forwarded via the switching fabric.
 Control packets (for example, packets carrying routing protocol
information) are forwarded from an input port to the routing
processor.
 Switching fabric
14 It is the heart of the router, it can be used to
forward data from incoming port to outgoing
port. Switching fabric can be implemented in
the following ways: Switching via memory,
Switching via bus and Switching via
interconnection network (Crossbar).
15 Output ports
 An output port stores packets received from the
switching fabric and transmits these packets on the
outgoing link by performing the necessary link-layer
and physical-layer functions.
 When a link is bidirectional, an output port will
typically be paired with the input port for that link on
the same line card (a printed circuit board containing
one or more input ports, which is connected to the
switching fabric).
 Scheduling discipline chooses among queued datagrams
for transmission by considering the issues like:
 Ordering (First Come First Served), Quality of Service (QoS),
fairness (protecting monopoly of resources).
16 Routing Processor

 The routing processor:

 Executes the routing protocols (which we’ll study in
the next classes and demonstrated in the laboratory
sessions),
 Maintains routing tables (by sharing routing table
information among routers) and attached link state
information (dependent on the routing algorithm
used), and
 Computes the forwarding table for the router
 It also performs the network management functions.
17 Internet protocol and IP addressing
 IP is part of the TCP/IP suite and is the most widely used
internetworking protocol.
 As with any protocol standard, IP is specified in two parts:
 The interface with a higher layer (TCP/UDP), specifying
the services that IP provides
 The actual protocol format and mechanisms
 IP basic characteristics:
 Connectionless - No connection is established before
sending data packets.
 Best Effort (unreliable) - No overhead is used to guarantee
packet delivery.
 Media Independent - Operates independently of the
medium carrying the data.
18
IP Header
19
IP header cont…..
 Version (4 bits): Indicates version number, to allow evolution of the
protocol; the value is 4.
 Header Length (4 bits): Length of header in 32-bit words. The
minimum value is five, for a minimum header length of 20 octets.
 Type of service: these bits were included in the IPv4 header to
allow different types of IP packets (reliability, precedence, delay,
and throughput parameters) to be distinguished from each other.
 Datagram Length (16 bits): it is total length of the datagram,
including header plus data, in octets (mostly less than or equal to
1500 bytes).
 Identifier, flags, fragmentation offset: these three fields have to do
with so-called IP fragmentation. The length of the fragment
(packet) is always dependent on the data link frame size (i.e.
Maximum Transmission Unit (MTU) of the current data link
protocol).
 Time-to-live (TTL): field is included to ensure that datagrams do
not circulate forever in the network. This field is decremented by
one each time the datagram is processed by a router. If the TTL
 IP header cont……
20  Protocol (8 bits): Indicates the next higher level protocol that is to receive
the data field at the destination; thus, this field identifies the type of the
next header in the packet after the IP header. Example values are TCP =
6; UDP = 17 ; ICMP = 1.
 Header Checksum: An error-detecting code applied to the header only.
 Source and Destination IP Address: Coded to allow a variable allocation of
bits to specify the network and the end system attached to the specified
network.
 The options parameter: it allows for future extensibility and for inclusion
of parameters that are usually not invoked. The currently defined options
are as follows:
 Security, source routing, route recording, stream identification and
time-stamping.
 Padding (variable): Used to ensure that the datagram header is a multiple
of 32 bits in length.
 Data (variable):
 IPv4 Addressing
21
 IP address: it is a 32-bit identifier for host and router interfaces (both
physical and logical interfaces have an ip address).
 Interface is a boundary between the host and the physical link in
computer networking.
 An IP address is technically associated with an interface, rather than with
the host or router containing that interface.
 These addresses are typically written in so-called dotted-decimal notation,
in which each byte of the address is written in its decimal form and is
separated by a period (dot) from other bytes in the address.
Example:
 These addresses cannot be chosen in a willy-nilly manner, however. A
portion of an interface’s IP address will be determined by the subnet to
which it is connected.
22 IPv4 Addressing cont…
23
IPv4 Addressing cont….
 Network Classes: the address is coded to allow a variable
allocation of bits to specify network and host. This encoding
provides flexibility in assigning addresses to hosts and allows
a mix of network sizes on an internet.
 The three principal network classes are best suited to the
following conditions:
 Class A: Few networks, each with many hosts
 Class B: Medium number of networks, each with a
medium number of hosts
 Class C: Many networks, each with a few hosts
 IPv4 Addressing cont…..
24

 The Internet’s address assignment strategy is known as Classless

Interdomain Routing (CIDR).
 CIDR generalizes the notion of subnet addressing. As with subnet
addressing, the 32-bit IP address is divided into two parts and again has
the dotted-decimal form a.b.c.d/x (x is a network prefix).
 The x most significant bits of an address of the form a.b.c.d/x constitute
the network portion of the IP address.
 The number of bits used in host portion determines the number of hosts
that we can have within the network.
 Types of Addresses in IPv4 Network
 Network address ‐ The address by which we refer to the network
 Broadcast address ‐ A special address used to send data to all hosts in
the network
 Host addresses ‐ The addresses assigned to the end devices (including
router interfaces) in the network
 Subnet and Network Prefixes (Subnet
25
Masking)
 Subnetting is the process of splitting the larger network into
a number of smaller networks.
 Within the subnetted network, the local routers must route
on the basis of an extended network number consisting of the
network portion of the IP address and the subnet number.
 Because of Subnetting we get the following advantages:
 Network traffic was reduced
 Network performance become optimized
 Network management become simple
 Network supports large geographic area
 The prefix length (x in the previous slide) is the number of
bits in the address that gives us the network portion.
 Subnet cont…
26
Example
27 Subnet cont…
How to Create Subnets
 To create a subnet answer the following questions:
 How many subnets does the chosen subnet mask
produced? It can be calculated using 2n-dp, where n is
the given network prefix and dp is the default prefix
for the given class (i.e. class A, B or C) when IP
address is a Classful IP or dp is the network prefix
given from ISP.
 How many valid hosts per subnet are available? It can
be calculated using 232-n-2.
 What are the valid subnets addresses?
 What’s the broadcast address of each subnets?
 What are the valid hosts in each subnet?
28 Subnet cont….
Example 1. consider the following class C Ip-address:
[Link]/25 ([Link])
[Link] = Network address
[Link] = Subnet mask (i.e.
11111111.11111111.11111111.10000000)
 How many subnets? Since 128 is 1 bit exceeded from
the default 24 bits on (10000000), the answer would
be 225-24=21=2.
 How many hosts per subnet? We have 7 (32-25) host
bits off (10000000), so the equation would be 27-2 =
126 hosts.
 What are the valid subnets? 256-128 = 128. Remember,
we’ll start at zero and count in our block size, so our
subnets are 0, 128.
 29 Subnet cont…..
Example 1. cont……
 What is the broadcast address for each subnet? The
number right before the next subnet address equals the
broadcast address. For the zero subnet 127 and for 128
subnet 255 are a broadcast addresses.
 W h a t a re t h e v a lid Subnet
h o s t s ?One
T h es Subnet
e a re tTwo
h e nu m be r s
between the subnet and broadcast
Subnet 0 address.
128
First host 1 129
Last host 126 254
Broadcast 127 255
30 Subnet cont….

Example 2: [Link]/20 ([Link])

class B network
Example 3: [Link]/10 ([Link]) class A
network
Example 4: [Link]/23 ([Link])
in a classless Ip-address with /19 prefix.
Example 5: [Link]/27
([Link]) in a classless IP-address
with /22 prefix.
31
Subnet cont…..
Variable Length Subnet Masks (VLSM)
 32 Reserved IPv4 Address Ranges
 Experimental Addresses: it ranges from [Link] to
[Link]. Currently, these addresses are listed as reserved
for future use.
 Multicast Addresses: it ranges [Link] to [Link]. It is
also subdivided into two types of addresses: reserved link local
addresses and globally scoped addresses.
 The IPv4 multicast addresses [Link] to [Link] are
reserved link local addresses. These addresses are to be used for
multicast groups on a local network. Packets to these destinations
are always transmitted with a time‐ to‐ live (TTL) value of 1.
 The globally scoped addresses are [Link] to [Link].
They may be used to multicast data across the Internet. For
example, [Link] has been reserved for Network Time Protocol
(NTP) to synchronize the time‐ of‐ day clocks of network devices.
33 Reserved IPv4 Address Range cont….
 Private Addresses: the address blocks belongs to this
category are:
 [Link] to [Link] ([Link] /8)
 [Link] to [Link] ([Link] /12)
 [Link] to [Link] ([Link] /16)
 Default Route: it is used as a "catch all" route when a
more specific route is not available ([Link]/8).
 Broadcast Address ([Link]/32)
 Loopback ([Link]/8) : the loopback is a special
address that hosts use to direct traffic to themselves.
 34 Reserved IPv4 Address Range cont
 Link‐ Local Addresses (Automatic Private IP Addressing): the
address block [Link] to [Link]
([Link] /16) are designated as link‐ local addresses
(private DHCP address range). These addresses can be
automatically assigned to the local host by the operating
system in environments where no IP configuration is available
(both static and dynamic IP address assignments).
NOTE: When the device get ip-address either statically or using
a DHCP server link local address is automatically replaced by
this new address.
 TEST‐ NET Addresses : the address block [Link] to
[Link] ([Link] /24) is set aside for teaching and
learning purposes. These addresses can be used in
documentation and network examples. However, network
35
IPv6: Structure and Address Space
 The IP (IPv4) has been the foundation of the Internet
and virtually all multivendor private internetworks.
 This protocol is reaching the end of its useful life and
a new protocol, known as IPv6 (IP version 6), has
been defined to ultimately replace IP.
 The driving motivation for the adoption of a new
version of IP was the limitation imposed by the 32-
bit address field in IPv4.
 With a 32-bit address field, it is possible in principle
to assign 2 32 different addresses, which is over 4
billion possible addresses.
36 IPv6 cont……
 Reasons for the inadequacy of 32-bit addresses
include the following:
 The two-level structure of the IP address is
convenient but wasteful of the address space.
 Networks are multiplying rapidly.
 Growth of TCP/IP usage into new areas will
result in a rapid growth in the demand for
unique IP addresses.
 Multiple IP addresses are required for a single
host.
 IPv6 cont……
37
 IPv6 includes the following enhancements over IPv4:
 Expanded address space: IPv6 uses 128-bit addresses.
 Improved option mechanism: IPv6 options are placed in
separate optional headers that are located between the
IPv6 header and the transport-layer header. It also
makes it easier to add additional options.
 Increased addressing flexibility: IPv6 includes the concept
of an anycast address in addition to the existing casting
mechanisms in IPv4 (i.e. Unicast, Multicast and
Broadcast).
 Support for resource allocation: IPv6 enables the labeling
of packets belonging to a particular traffic flow for
which the sender requests special handling. This aids in
 IPv6 Header
38

 The following fields are defined in IPv6:

 Version: this 4-bit field identifies the IP version number. IPv6
carries a value of 6 in this field.
 Traffic class: this 8-bit field is similar in spirit to the TOS field
we saw in IPv4.
 Flow label: this 20-bit field is used to identify a flow (the
datagram that needs special treatments due to different
reasons) of datagrams.
39 IPv6 Header cont……
 Payload length (16 bits): the number of bytes in the IPv6 datagram
following the fixed-length datagram header.
 Next header: transport protocol to which the contents of this datagram
will be delivered.
 Hop limit: the total number of routers in transmission. If the count
reaches zero, the datagram is discarded.
 Source and destination addresses
 Data
 Several fields appearing in the IPv4 datagram are no longer present in
the IPv6 datagram:
 Fragmentation/Reassembly: IPv6 does not allow for fragmentation and
reassembly at intermediate routers. If an IPv6 datagram received by a
router is too large to be forwarded over the outgoing link, the router
simply drops the datagram and sends a “Packet Too Big” ICMP error
message back to the sender.
 Header checksum
 Options
 IPv6 Addresses
40
 IPv6 addresses are 128 bits in length. Addresses are assigned to
individual interfaces on nodes, not to the nodes themselves.
 IPv6 allows three types of addresses:
 Unicast: An identifier for a single interface.
 Anycast: An identifier for a set of interfaces. A packet sent to
an anycast address is delivered to one of the interfaces
identified by that address (the “nearest” one, according to
the routing protocols’ measure of distance).
 Multicast: An identifier for a set of interfaces.
 IPv6 Addresses cont.….
41
 IPv6 addresses are represented by treating the 128-bit address as a
sequence of 8 16-bit numbers, and representing this in the form of
eight hexadecimal numbers divided by colons, for example:
[Link]
 One to three zeroes that appear as the leading digits in any colon-
delimited hexadecimal grouping may be dropped
([Link])
 A group of all zeroes, or consecutive groups of all zeroes, can be
substituted by a double colon, but this may only be done once in an
address. Example: [Link] or
[Link]
Structure of IPv6
ipv6-address/prefix-length
 prefix-length is a decimal value specifying how many of the leftmost
 Internet Control Message Protocol (ICMP)
42

 ICMP provides a means for transferring messages

from routers and hosts each other. In essence, ICMP
provides feedback about problems in the
communication environment.
Examples: when a datagram cannot reach its
destination, or when the router does not have the
buffering capacity to forward a datagram.
 An ICMP message is sent in response to a datagram,
either by a router along the datagram’s path or by
the intended destination host.
 Although ICMP is, in effect, at the same level as IP
in the TCP/IP architecture, it is a user of IP.
 ICMP cont….
43

 An ICMP message starts with a 64-bit header consisting of the

following:
 Type (8 bits): Specifies the type of ICMP message.
 Code (8 bits): Used to specify parameters of the message that
can be encoded in one or a few bits.
 Checksum (16 bits): Checksum of the entire ICMP message.
 Parameters (32 bits): Used to specify more lengthy parameters.
 ICMP cont….
44  ICMP error report kinds:
 The destination unreachable message covers a number of contingencies.
 A router will return a time exceeded message if the lifetime of the
datagram expires.
 A syntactic or semantic error in an IP header will cause a parameter
problem message.
 The echo and echo reply messages (pinging and replaying the ping
message)provide a mechanism for testing that communication is
possible between entities.
45 ICMP cont.…..
 The source quench message provides a rudimentary
form of flow control. Either a router or a destination
host may send this message to a source host, requesting
that it reduce the rate at which it is sending traffic to
the internet destination.
 A router sends a redirect message to a host on a
directly connected router to advise the host of a better
route to a particular destination.
 The timestamp and timestamp reply messages provide
a mechanism for sampling the delay characteristics of
the internet.
 The address mask request and address mask reply
messages are useful in an environment that includes
 ICMPv6
46  ICMPv6 is used by IPv6 nodes to report errors encountered in
processing packets, and to perform other internet-layer functions,
such as diagnostics (ICMPv6 "ping"). ICMPv6 is an integral part of
IPv6 and MUST be fully implemented by every IPv6 node.
 ICMPv6 messages are grouped into two classes: error messages and
informational messages.

ICMPv6 error messages ICMPv6 informational

messages:
1 Destination 128 Echo Request
Unreachable
2 Packet Too Big 129 Echo Reply
3 Time Exceeded
47
ICMPv6 cont…..
 The ICMPv6 packet consists of a header and the
protocol payload. The header contains only three
fields: type (8 bits), code (8 bits), and checksum (16
bits).
 Type specifies the type of the message. Values in the
range from 0 to 127 (high-order bit is 0) indicate
an error message, while values in the range from
1 2 8 t o 2 5 5 (h ig h -o rder bi t i s 1 ) i n d i c a t e a n
information message.
 The code field value depends on the message type
and provides an additional level of message
granularity.
48
Address Mapping
 Logical address (IP address): it is a network address uniquely
identify networks in the wide area network.
 Physical address: it is a MAC (Medium Access Control) address,
which provides a physical address for a host port attached to
the LAN.
 To deliver an IP datagram to a destination host, a mapping
must be made from the IP address to the subnetwork (MAC)
address for that last hop.
 If a datagram traverses one or more routers between source
and destination hosts, then the mapping must be done in the
final router, which is attached to the same subnetwork as the
destination host.
 If a datagram is sent from one host to another on the same
subnetwork, then the source host must do the mapping.
 Address Mapping cont….
49
 For this purpose, a number of approaches are possible, it
include:
 Each system can maintain a local table of IP addresses and
matching subnetwork addresses for possible correspondents.
 A centralized directory can be maintained on each
subnetwork that contains the IP-subnet address mappings.
 An address resolution protocol can be used.
 Address Resolution Protocol (ARP)
50  ARP allows dynamic distribution of the information needed to
build tables to translate an IP address A into a 48-bit
Ethernet address; the protocol can be used for any broadcast
network. ARP exploits the broadcast property of a LAN.
 ARP works as follows:
1. Each system on the LAN maintains a table of known IP-
subnetwork address mappings.
2. When a subnetwork address is needed for an IP address, and
the mapping is not found in the system’s table, the system
uses ARP directly on top of the LAN protocol (e.g., IEEE
802) to broadcast a request. The broadcast message contains
the IP address for which a subnetwork address is needed.
3. Other hosts on the subnetwork listen for ARP messages and
reply when a match occurs. The reply includes both the IP
and subnetwork addresses of the requested and replying host.
 Internet Group Management Protocol (IGMP)
51  IGMP interaction is limited to a host and its attached router, another protocol
is required to coordinate the multicast routers throughout the Internet, so that
multicast datagrams are routed to their final destinations. This latter
functionality is accomplished by network-layer multicast routing algorithms.
Network-layer multicast in the Internet thus consists of two complementary
components: IGMP and multicast routing protocols.
 IGMP has only three message types. IGMP messages are encapsulated within an
IP datagram, with an IP protocol number of 2.
 The membership_query message is sent by a router to all hosts on an
attached interface to determine the set of all multicast groups that have
been joined by the hosts on that interface.
 membership_report messages can be used by the host to replay the router
message or joins a multicast group without waiting for a membership_query
message.
 The final type of IGMP message is the leave_group message (this message is
optional).
 How does a router detect the host leaves the group? The router infers that a
host is no longer in the multicast group if it no longer responds to a
membership_query message with the given group address.
Example:
52
Network Address Translation (NAT)
 NAT cont….
53
 The basic idea behind NAT is for the ISP to assign each
home or business a single (few) public IP address(es) for
Internet traffic.
 Within the customer network, every computer gets a
unique IP address, which is used for routing in the
intranet traffic. However, just before a packet exits the
customer network and goes to the ISP, an address
translation from the unique internal IP address to the
shared public IP address takes place.
 This translation makes use of three ranges of IP
addresses that have been declared as private. The only
rule is that no packets containing these addresses may
appear on the Internet itself.
 If all datagrams arriving at the NAT router from the
WAN have the same destination IP address, then how
does the router know the internal host to which it should
forward a given datagram?
54 NAT cont….
 NAT has enjoyed widespread deployment in recent years.
But it has some limitations:
 First port numbers are meant to be used for addressing
processes, not for addressing hosts.
 Second routers are supposed to process packets only up
to layer 3.
 Third the NAT protocol violates the so-called end-to-
end argument; that is, hosts should be talking directly
with each other, without interfering nodes modifying IP
addresses and port numbers.
 Fourth it interferes with P2P applications, including
P2P file-sharing applications and P2P Voice-over-IP
 Routing Algorithm
55
 It is that part of the network layer software responsible
for deciding which output line an incoming packet should
be transmitted on.
 If the network uses datagrams internally, this decision
must be made a new path for every arriving data packet.
 If the network uses virtual circuits internally, routing
decisions are made only when a new virtual circuit
(sometimes called session) is being set up.
 A host is attached directly to one router, the default
router for the host.
 The purpose of a routing algorithm is then simple: given
a set of routers, with links connecting the routers, a
routing algorithm finds a “good” path from source
router to destination router. Typically, a good path is
one that has the least cost.
56 Routing Algorithm cont….
 The design of routing algorithm may have the following
properties into consideration: correctness, simplicity,
robustness, stability, fairness, and efficiency.
 Routing algorithms can be grouped into two major
classes: non-adaptive and adaptive.
 Non-adaptive (static) algorithms
 Adaptive (dynamic) algorithms
 These dynamic routing algorithms differ in where they get
their information, when they change the routes , and what
metric is used for optimization.
 Others classify routing algorithms into: global or
decentralized, link-state or distance-vector and load
sensitive or load-insensitive.
 Routing Algorithm cont….
57
The Optimality Principle
 It states that if router J is on the optimal path from router I
to router K, then the optimal path from J to K also falls
along the same route.
 To see this, call the part of the route from I to J r1 and the
rest of the route r2.
 As a direct consequence of the optimality principle, we can
see that the set of optimal routes from all sources to a given
destination form a tree rooted at the destination. Such a tree
is called a sink tree and is illustrated in the next slide.
58 Shortest Path Algorithm
 The idea is to build a graph of the network, with
each node of the graph representing a router and
each edge of the graph representing a
communication line, or link.
 To choose a route between a given pair of routers,
the algorithm just finds the shortest path between
them on the graph.
 The concept of a shortest path deserves some
explanation.
 One way of measuring path length is the number of
hops.
 Another metric is the geographic distance in kilometers.
59 Flooding
 In this algorithm every incoming packet is sent out on
every outgoing line except the one it arrived on.
 Flooding obviously generates vast numbers of duplicate
packets, in fact, an infinite number unless some measures
are taken to damp the process.
 One such measure is to have a hop counter contained in the
header of each packet that is decremented at each hop, with
the packet being discarded when the counter reaches zero.
 Another technique for damming the flood is to have routers
keep track of which packets have been flooded, to avoid
sending them out a second time.
 It is effective for broadcasting information, tremendously
60 Flooding cont….
61 Hierarchical Routing
 As networks grow in size, the router routing tables grow
proportionally. It affects router memory, CPU time and
bandwidth to send status reports to others.
 At a certain point, it is no longer feasible for every
router to have an entry for every other router, so the
routing will have to be done hierarchically.
 When hierarchical routing is used, the routers are
divided regions.
 Each router knows all the details about how to route
packets to destinations within its own region but knows
nothing about other regions.
 When a single network becomes very large, an
interesting question is ‘‘how many levels should the
hierarchy have?’’
62 Distance Vector Routing
 A distance vector routing algorithm operates by having each
router maintain a table (i.e., a vector) giving the best known
distance to each destination and which link to use to get
there.
 These tables are updated by exchanging information with the
neighbors.
 In distance vector routing, each router maintains a routing
table indexed by, and containing one entry for each router in
the network.
 This entry has two parts:
 The preferred outgoing line to use for that destination and
 An estimate of the distance to that destination.
 The commonly used distance vector routing protocols are: RIP
(Routing Information Protocol), BGP (Border Gateway
 Distance Vector Routing cont….
63
64 Link State Routing

 The idea behind link state routing can be stated as

five parts. Each router must do the following things
to make it work:
1. Discover its neighbors and learn their network
addresses.
2. Set the distance or cost metric to each of its
neighbors.
3. Construct a packet telling all it has just learned.
4. Send this packet to and receive packets from all
other routers.
5. Compute the shortest path to every other router.
65
Link state cont….
66 Link State Routing cont…..
Learning about the Neighbors
 When a router is booted, its first task is
to learn who its neighbors are.
 It accomplishes this goal by sending a
special HELLO packet on each point-to-
point line. The router on the other end is
expected to send back a reply giving its
name.
67 Link State Routing cont…..
Setting Link Costs
 The link state routing algorithm requires each
link to have a distance or cost metric for finding
shortest paths.
 The cost to reach neighbors can be set
automatically, or configured by the network
operator.
 A common choice is to make the cost inversely
proportional to the bandwidth and distance gap
between two routers of the link.
 The most direct way to determine this delay is
to send over the line a special ECHO packet that
the other side is required to send back
immediately.
 By measuring the round-trip time and dividing
it by two, the sending router can get a
reasonable estimate of the delay.
 Link State Routing cont…..
68 Building Link State Packets
 Building the link state packets is easy. The hard
part is determining when to build them.
 One possibility is to build them periodically, that is,
at regular intervals.
 Another possibility is to build them when some
significant event occurs, such as a line or neighbor
going down or coming back up again or changing
its properties appreciably.
69 Link State Routing cont…..
Distributing the Link State Packets
 The trickiest part of the algorithm is distributing the link
state packets.
 All of the routers must get all of the link state packets
quickly and reliably.
 If different routers are using different versions of the
topology, the routes they compute can have
inconsistencies such as loops, unreachable machines, and
other problems.
 The fundamental idea is to use flooding to distribute the
link state packets to all routers.
70 Link State Routing cont…..
Computing the New Routes
 Once a router has accumulated a full set of link state
packets, it can construct the entire network graph
because every link is represented.
 Every link is, in fact, represented twice, once for
each direction. The different directions may even
have different costs.
 The shortest-path computations may then find
different paths from router A to B than from router
B to A.
 71 Link State Routing cont…..
 Compared to distance vector routing, link state routing
requires more memory and computation.
 For a network with n routers, each of which has k
neighbors, the memory required to store the input data is
proportional to kn, which is at least as large as a routing
table listing all the destinations.
 Also, the computation time grows faster than kn, even
with the most efficient data structures, an issue in large
networks.
 Nevertheless, in many practical situations, link state
routing works well because it does not suffer from slow
convergence problems.
 Asynchronous Transfer Mode (ATM)
72
 Asynchronous transfer mode is a switching and multiplexing
technology that employs small, fixed-length packets called cells.
 A fixed-size packet makes switching and multiplexing efficient
and a small cell size was chosen to support delay-intolerant
interactive voice service.
 ATM is a connection-oriented packet-switching technology that
was designed to provide the performance of a circuit-switching
network and the flexibility and efficiency of a packet-switching
network.
 ATM was intended to provide a unified networking standard
for both circuit-switched and packet-switched traffic, and to
support data, voice, and video with appropriate QoS
mechanisms.
 ATM cont…..
73
ATM Logical Connections
 Virtual channel connections (VCCs) is analogous to a virtual
circuit; it is the basic unit of switching in an ATM network.
 A VCC is set up between two end users through the network,
and a variable-rate, full-duplex flow of fixed-size cells is
exchanged over the connection. VCCs are also used for
user–network exchange (control signaling) and
network–network exchange (network management and
routing).
 A virtual path connection (VPC) is a bundle of VCCs that have
the same endpoints. Thus, all of the cells flowing over all of
the VCCs in a single VPC are switched together.
 ATM cont….
74
 Advantages of using a virtual paths:
 Simplified network architecture
 Increased network performance and reliability
 Reduced processing and short connection setup time
 Enhanced network services
 Virtual Path/Virtual Channel Characteristics ITU-T
recommend the following as characteristics:
 Quality of service (QoS)
 Switched and semipermanent virtual channel connections
 Cell sequence integrity
 Traffic parameter negotiation and usage monitoring
 Virtual channel identifier restriction within a VPC (unique for
 ATM Cell Format
75
76
ATM Cell Format cont…
 The Generic Flow Control (GFC): it can be used for control of cell
flow only at the local user–network interface. The field could be
used to assist the customer in controlling the flow of traffic for
different qualities of service. In any case, the GFC mechanism is
used to alleviate short-term overload conditions in the network.
 The virtual path identifier (VPI) constitutes a routing field for the
network. In the network cell allows support for an expanded
number of VPCs internal to the network, to include those
supporting subscribers and those required for network management.
The virtual channel identifier (VCI) is used for routing to and from
the end user.
 The Payload Type (PT) field indicates the type of information in
the information field (the detail codes found in the next slide).
 The Cell Loss Priority (CLP) bit is used to provide guidance to the
network in the event of congestion. A value of 0 indicates a cell of
relatively higher priority, which should not be discarded unless no
other alternative is available. A value of 1 indicates that this cell is
subject to discard within the network.
 The Header Error Control (HEC) field is an 8-bit error code that
77 ATM Cell Format cont…..
 Next Chapter IV
Transport Layer
78