Network Layer Delivery,Forwarding and Routing

Lecture-17

Computer Communication Networking

(University of central Punjab)
 Lecture Topics
 Introduction, the Network Layer

 What is routing

 Internet and autonomous systems

 Types of routing algorithms and routing methods

 Static - Shortest Path Routing

 Introduction – The Network Layer

 Network layer is concerned with getting packets from

the source to destination

 Network layer must know the topology of the subnet, it must take care to
choose routes to avoid overloading
 Network layer – Connectionless Services

 Connectionless services – datagram subnet

 Datagram subnet – Router maintains table of entries

 Entries - pair consisting of a destination and directly connected outgoing line to

use for that destination

 Router can route packets on any route in datagram subnet

 Network layer – Connection oriented services
 Connection oriented services, Virtual circuit (VC) subnet

 VC subnet avoids choosing new routes for every packet but use the
same route stored in router tables

 VC is only there for the session and is terminated at the end of

connection

 Packets are routed based on VC identifiers

 Functions of Network layer
application
transport
network
 Transfer segment from sending to receiving host data link network
physical data link
network network
data link physical data link
physical physical
 On sending side encapsulates segments into datagrams network
data link
physical network
data link
 On physical
receiver side, delivers segments to transport layer
network
network data link
data link physical
physical
 Network layer protocols in every host, router network
data link application
physical transport
network
 Router examines header fields in all IP datagrams passing through it data link
physical
 Key Network-Layer Functions
Analogy:
 Forwarding

 Move packets from router’s input to

• Routing: Process of planning
 appropriate router output
trip from source to
destination
 Routing

 Routing algorithms • Forwarding: Process of

 Determine route taken by packets getting through single
from source to destination interchange
 Network Layer Connection and Connection-Less
Services
 3rd important function
 Connection Less Network Service
 Datagram Network

 Connection Oriented Network Service

 Virtual Circuit Network

 Datagram Network provides network-layer connectionless service

 VC Network provides network-layer connection service

( Source: Tannenbaum Book )
 Virtual Circuits
“Source-to-Destination Path behaves much like telephone circuit”
• Performance-wise
• Network actions along Source-to-Destination path

 Call Setup, Teardown for each call before data can flow
 Each Packet carries VC identifier (not destination host address)
 Every router on source-destination path maintains “state” for each passing
connection
 Link, Router Resources (bandwidth, buffers) may be allocated to VC
 Virtual Circuits (Signaling Protocols)

 Used to setup, maintain teardown VC

 Used in ATM, frame-relay, X.25
 Not used in today’s Internet
 Datagram Networks

 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-destination pair may take different paths
 What is Routing?

 Routing is all about network layer design

 The algorithms that manages the routing tables and makes routing decisions is
called routing algorithm

 Routing algorithm is the main topic of interest in network layer design (more
about routing will follow on coming slides)
 Routing….. 5
3
 “good” path: B C 5
2
 typically means minimum cost path A 2
3
1 F
 link “cost” : 1 2
D E
 Weight on a graph edge reflect “cost” of traversing the “link” 1

 Can based on different performance criteria

 E.g.,

 delay,
NOTE:
 £, cost,
Resulting shortest path
 Hop count (weight = 1)
may not have fewest hops.
 Throughput (utilization)
 or congestion level
 Types of Routing Algorithms

 Non-adaptive routing algorithms (a.k.a. static routing algorithms). Choice of

route is computed in advance, offline and downloaded to the routers during
the network boot

 Adaptive-routing algorithms (a.k.a. dynamic routing algorithms). Routing

decisions can be changed to reflect changes in topology and traffic load
 Routing Methods

 Next-Hop Routing
 The routing table holds only the address of the next hop instead of holding info
about the complete route.
 A technique to reduce the routing table

 Network-Specific Routing
 Insteadof having an entry for every host connected to the same physical
network, only one entry is used to give the network address.
 Another way to reducing the routing table.
 Routing Methods…..

 Host-Specific Routing
 The destination host address is given in the routing table.
 Useful when the administrator wants to have more control over the routing to a
particular host.
 Default Routing
 A routing method in which a router is assigned to receive all packets with no
match in the routing table.
Next-Hop Routing

( Source: TCP/IP Protocol Suite, by Forouzan )

Network-Specific Routing

( Source: TCP/IP Protocol Suite, by Forouzan )

Host-Specific
Routing
( Source: TCP/IP
Protocol Suite,
by Forouzan )
Default Routing

( Source: TCP/IP Protocol

Suite, by Forouzan )
 Some Static and Dynamic routing algorithms

 Non-adaptive / Static routing algorithms

 Shortestpath routing
 Flooding
 Dynamic / Adaptive routing algorithms
 Distance Vector Routing
 Link State Routing
 Pre-requisite - Terms and definitions
 Shortest Path
 thepath with the least cost if there are multiple paths available between two
routers.
 Path (Link) Cost
 the cost of transmitting a packet over a path (link),
 canbe manually configured, or can be based on various parameters (i.e.,
metrics) such as the bandwidth availability or congestion.
 adjusting the cost influences the results produced by routing algorithms.
 Pre-requisite - Terms and definitions …..

 Routing Metric
 Used to decide the cost value associated with each link,
 can be based on a number of functions – bandwidth, distance etc.,
 Capacity, Delay, Expense and Error Rate are examples of routing metrics.

 Hops
 Every time a packet is received by a router (on the route to the destination), this is known as a
hop
 Hop Count
 the total number of hops a packet needs to make before arriving at its destination.
 Pre-requisite - Terms and definitions …..

 Optimality Principle
 R1  R2  R3
 If router R2 is on the optimal path from router R1 to router R3, the optimal path
from R2 to R3 also falls along the same route.
 Next Hop Routing
 Given the Optimality principle, a router need only know the next router to
transmit to, and need not necessarily have a global view of the network
 Shortest Path Routing

 Subnet represented by a graph, each node of the graph represents

router and each arc of graph represents a communication link
 Shortest path between two nodes of the graph can be computed by
several algorithms (e.g.Dijkstra)
 Path length (shortest path) can be computed as a function of,
 Numberof hops, Geographic distance, Bandwidth, mean queue length,
measured delay etc
 Static Routing--Review

 Manually performed by the network administrator. The administrator is

responsible for discovering and propagating routes through the network

 A device once configured, simply forwards packets out the predetermined

ports. There is no communication between routers regarding the current
topology of the network
 Static Routing….
 Advantages:
 Manual definition of a default route
 Definition of route not automatically advertised
 To provide more secure network environment

 Disadvantages:
 Static routes require considerable amount of coordination and maintenance in
non-trivial network environments
 Static routes can not dynamically adapt to the current operational state of
network
 Dijkstra Algorithm

 network topology, link costs known to all nodes

 accomplished via “link state broadcast”
 all nodes have same info
 calculates least cost paths from one node (‘source”) to all other nodes
 produces routing table for that node
 iterative: after k iterations, know least cost path to k destinations
 Dijsktra Algorithm…..
Notation:
 c(i,j):

link cost from node i to j. cost infinite if not direct neighbors

 D(v):

current value of cost of path from source to destination v

 p(v):

predecessor node along path from source to v

 N:

set of nodes whose least cost paths have become known

 Dijsktra Algorithm…..
1 Initialization:
2 N = {A}
3 for all nodes v
4 if v adjacent to A
5 then D(v) = c(A,v)
6 else D(v) = infinity
7
8 Loop
Assuming A is a source node,
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
 Dijsktra Algorithm…..

3
B C
2 5

A 2 1 F
3
1
2
D E
1
 Dijkstra’s Algorithm: an example
Step start N D(B),p(B) D(C),p(C) D(D),p(D) D(E),p(E) D(F),p(F)
0 A 2,A 5,A 1,A infinity infinity
1 AD 2,A 4,D 2,D infinity
2 ADE 2,A 3,E 4,E
3 ADEB 3,E 4,E
4 ADEBC 4,E
5 ADEBCF

5
3
B C 5
2
A 2 1 F
3
1 2
D E
1
 Distance Vector Routing
 Itallows each device in the network to automatically build and maintain a local
IP routing table
 Eachrouter in the internetwork maintains a distance or cost from itself to every
known destination, this value represents the overall desirability of the path
 The path represented by the smallest cost becomes the preferred path to reach
destination.
 The information is maintained in a distance vector table, which is periodically
advertised to each neighbouring router
 Distance Vector Routing…..
 When a node is first turned on, it constructs its initial distance table (see below
for how to construct) and then transmits the table to its neighbours.
 On receipt of a packet containing the distance table from a neighbour, a node
compares the entries in this table with their own. If an entry contains a lower
cost, it adds the new entry to its table.
 In this way, the entries are updated and created across the network.
 Thisrouting algorithm is known with these names: Bellman-Ford, Ford-
Fulkerson
 Distance Vector Routing…..
Distance Table data structure:
 Each node has its own distance table
 Each row for each possible destination
 Each column for each directly-attached neighbor to node
 example: in node X, for destination Y via neighbor Z

distance from X to
X = Y, via Z as next hop
D (Y,Z)
Z
= c(X,Z) + minw{D (Y,w)}
 Link State Routing

 Algorithms use principle of link state to determine network topology

 A linkstate is the description of an interface on a router (e.g. IP address, subnet

mask) and its relationship to the neighbouring routers

 The collection of these link states forms a link state database

 Link State Routing, determining network topology

 Each router identifies all other routing devices on the directly connected
networks

 Each router advertises a list of all directly connected network links and
associated cost of each link through exchange of link state advertisements
(LSAs)

 Using LSAs, each router creates a database detailing the current topology
Link State Routing….
Link State Routing….
 Link State Routing….

( After R1 receives the link state info from R2 )

Link State Routing….
Link State Routing….
Link State Routing….
Link State Routing….
 Path Vector Routing
 Path vector somewhat similar to the distance vector algorithm

 Networks are advertised as destination addresses and path descriptions to reach

those destinations

 The name path vector routing –because a route is defined as a pairing between a
destination and the attributes of the path to that destination (routers receive a
vector that contains paths to a set of destinations)
 Recommended Reading

1. Chapters 6 and 13 of Forouzan’s Book (Behrouz A. Forouzan, TCP/IP Protocol Suite, 2/e,
McGraw Hall).
2. Sections 5.2 (5.2.1-5.2.7) and 5.6 (5.6.4-5.6.5) of Tanenbaum’s Book (Andrew S.
Tanenbaum, Computer Networks, 4/e, Prentice Hall, 2003, ISBN 0-13-038488-7)
3. Sections 4.2, 4.3 and 4.5 of Kurose’s Book (James F. Kurose, Keith W. Ross, Computer
Networking, 2/e, Addison-Wesley, ISBN 0-201-97699-4).
4. Sections 9.4, 10.2, 16.1 and Apendix 10A of Stallings’ book (William Stallings, Data &
Computer Communications, 6/e, Prentice Hall, 2000. ISBN: 0-13-084370-9).
5. Chapters 13 and 25 of Comer’s book (Douglas E. Comer, Computer Networks and
Internets with Internet Applications, 3/e, Prentice Hall, 2001, ISBN 0-13-091449-5).