1 of 15

Computer Networking: A Top Down Approach �6^th edition �Jim Kurose, Keith Ross�Addison-Wesley�March 2012

Computer Communications �& Networks

CSNC-2413

Core Network, Performance

Lec: 5

2 of 15

Chapter 1: roadmap

1.1 what is the Internet?

1.2 network edge

end systems, access networks, links

1.3 network core

circuit switching, packet switching, network structure

1.4 delay, loss, throughput in networks

1.5 protocol layers, service models

1.6 networks under attack: security

1.7 history

2

3 of 15

The network core

Mesh of interconnected routers
Fundamental question…

“how is data transferred thru N/W”

Circuit switching

resources along the path reserved �for Tx duration
e.g, telephone network
guaranteed service

Packet switching

N/W resources used on demand
e.g, Internet
best effort service

3

4 of 15

Circuit switching

Each link shown has four circuits

call gets 2^nd circuit in top link
call gets 1^st circuit in right link

dedicated resources, no sharing

performance guaranteed

circuit segment idle, if not used by call (no sharing)

4

Network establishes an end-to-end circuit between hosts
Resources reserved for the “call” (B/W, buffer space)

5 of 15

Multiplexing in Circuit switching

5

Circuits implemented on a link using… FDM (Freq Div Mux) or �TDM (Time Div Mux)

FDM

freq spectrum divided into freq bands
each connection allocated a freq band

TDM

time divided into frames
frames have fixed slots
each connection allocated a time slot in every frame

6 of 15

6

Time Division Mux (TDM)

7 of 15

7

TDM : DeMux

8 of 15

TDM circuit: data Tx rate

8

How to determine data Tx rates of TDM circuit/channel…
What is TDM circuit/channel…

Given…

Frame rate = 8000 frames/sec

Slot size = 8 bits

Data Rate = Frame Rate x Slot Size

Sol…

Data rate of one circuit = 8000 x 8 = 64 kbps

Data rate of TDM line = 8000 x (8 x 4) = 256 kbps

9 of 15

TDM circuit: file Tx time

9

How much time it takes to send a 640 kbits file from Host A to Host B over a circuit switched network. All links in network use TDM with 24 slots and have a line rate of 1.536 Mbps. Also, it takes 500 ms to establish the circuit before transmission can take place.

Given…

TDM line rate = 1.536 Mbps Frame size = 24 slots

Circuit establishment delay = 500 ms File size = 640 kbits

Sol…

Data rate of one TDM circuit = 1.536 Mpbs / 24 = 64 kbps

Tx time of file = 640 kbits / 64 kbps = 10 sec

Total File sending time = 500 ms + 10 sec = 10.5 sec

(actual file sending time would also involve propagation delay…)

10 of 15

The network core

Packet-switching:

hosts break application-layer �messages into smaller chunks, packets
pkts forwarded from one router to the next, across a set of links
each pkt uses full data rate of link
sources use network resources on demand

Bandwidth division into “pieces”

Dedicated allocation

Resource reservation

10

11 of 15

Multiplexing in Packet switching: Statistical Mux

Statistical Multiplexing:

on-demand resource allocation;
pkts of A & B do not have fixed time slots/sequence at output link

TDM had…

pre-allocated time slots; each host gets same slot in all TDM frames

A

B

C

10 Mb/s

Ethernet

1.5 Mb/s

D

E

statistical multiplexing

queue of packets

waiting for output

link

11

12 of 15

12

Statistical Mux vs TDM

Why do we need address here…??

In a synchronous time division multiplexer, many of the time slots in a frame are often wasted. Statistical time division multiplexing provides a generally more efficient service than synchronous TDM for the support of terminals. With statistical TDM, time slots are not preassigned to particular data sources. Rather, user data are buffered and transmitted as rapidly as possible using available time slots. As with a synchronous TDM, the statistical multiplexer has a number of I/O lines on one side and a higher speed multiplexed line on the other. Each I/O line has a buffer associated with it. In the case of the statistical multiplexer, there are n I/O lines, but only k, where k < n, time slots available on the TDM frame. For input, the function of the multiplexer is to scan the input buffers, collecting data until a frame is filled, and then send the frame. On output, the multiplexer receives a frame and distributes the slots of data to the appropriate output buffers. Because statistical TDM takes advantage of the fact that the attached devices are not all transmitting all of the time, the data rate on the multiplexed line is less than the sum of the data rates of the attached devices. Thus, a statistical multiplexer can use a lower data rate to support as many devices as a synchronous multiplexer. However there is more overhead per slot for statistical TDM because each slot carries an address as well as data. The difficulty with this approach is that, while the average aggregate input may be less than the multiplexed line capacity, there may be peak periods when the input exceeds capacity. The solution to this problem is to include a buffer in the multiplexer to hold temporary excess input. There is a tradeoff between the size of the buffer used and the data rate of the line. We would like to use the smallest possible buffer and the smallest possible data rate, but a reduction in one requires an increase in the other.

13 of 15

N users over 1 Mbps link
each Tx at 100 kbps when “active”; active 10% of time
Circuit-switching supports 10 users
Packet-switching allows more users because of Tx being intermittent

Packet switching may allow more users to use network…

……...

N users

1 Mbps link

Packet-switching Vs Circuit-switching

Packet switching may allow a user access to greater B/W…

In circuit-switching, user can only utilize his part of B/W

1/10 of B/W for 10 users

In packet-switching, with less number of users active at any given time, �a user may avail more B/W

13

14 of 15

Is packet switching a “ winner…?”

Great for bursty data

resource sharing
simpler, no call setup

Excessive congestion possible

pkt delay & loss
protocols needed for reliable data transfer, congestion control

Q: How to provide circuit-like behavior?

bandwidth guarantees needed for audio/video apps
still an unsolved problem (chapter 7)

14

Packet-switching Vs Circuit-switching

15 of 15

Two key network-core functions

Forwarding

move pkts from router’s input port �to appropriate output port

Routing

determine source-to-destination route taken by pkts

routing algorithms

routing algorithm

local forwarding table

header value

output link

0100

0101

0111

1001

3

2

1

2

3

0111

dest address in arriving

packet’s header

15

How do pkts make their way thru a pkt switched network…