Unit 6: Introduction Flashcards Preview

Computer Systems & Networks > Unit 6: Introduction > Flashcards

Flashcards in Unit 6: Introduction Deck (48):
1

Define bandwidth and unit it is measured in

Transmission rate of data - bits per second

2

What does a packet switch do?

Forwards packets (chunks) of data

3

What is the ‘nuts and bolts’ view of a network?

The internet is a network of networks i.e. consisting of interconnected ISPs. It consists of protocols that control sending and receiving of messages

4

What is the ‘service’ view of a network?

The internet is an infrastructure that provides services to applications and a programming interface to applications

5

Define a protocol

Protocols define the format and order of messages sent/received among network entities, and the actions taken on message transmission/receipt

6

What constitutes network edge?

The hosts - clients and servers

7

What constitutes access networks?

The wired and wireless communication links

8

What is the network core?

A mesh of interconnected routers

9

a) What does DSL stand for?
b) Describe a DSL access net (what is transmitted/where?)

a) DSL = Digital Subscriber Line
b) In DSL, voice and data are transmitted at different frequencies over a dedicated existing telephone line to a central office. The data goes to the internet, voice goes to the telephone net

10

What is frequency division multiplexing?

Where different channels are transmitted in different frequency bands

11

Describe a cable network access net

In a cable network a fibre cable attaches homes to an ISP router. Home share access networks to cable headend

12

State a difference between DSL and cable networks

Access networks are shared in cable networks, whereas DSL have dedicated access to the central office

13

What type of cable is used in cable networks?

Hybrid fiber coaxial (HFC) cables which are asymmetric (one direction)

14

Describe the modern home network access net and the different connection points

A wireless access point connects wireless devices to the router, in which wired devices are connected directly via wired ethernet. The router is connected to a cable or DSL modem.

15

Briefly describe wide-area wireless access

Cellular operators (e.g. Telco) provide 3G, 4G etc. by using technologies such as LTE

16

Define a bit in the context of physical media

What propagates between the transmitter and receiver pairs

17

Define a physical link in the context of physical media

What lies between a transmitter and receiver

18

Define guided media in the context of physical media

Signals that propagate in solid media e.g. copper

19

Define a unguided media in the context of physical media

Signals that propagate freely e.g. radio

20

Define a twisted pair in the context of physical media

Two insulated copper wires used in ethernet with high access speed

21

a) Define a coaxial cable in the context of physical media.
b) Where is it used?

a) Two concentric copper conductors which are
b) Broadband

22

a) Define a fiber optic cable in the context of physical media
b) Give and explain a benefit of fiber optic cables

A high speed glass fiber that carries light pulses (each represent a bit). It has a low error rate as it is immune to electromagnetic noise and repeaters are spaced far apart

23

a) Describe radio signals in the context of physical media
b) Give and explain a drawback of radio signals

Signals are carried in the electromagnetic spectrum. It is bidirectional. Radio is susceptible to the environment because of reflection, obstruction and interference

24

Give 4 types of radio link

Terrestrial microwave, LAN (wifi), wide-area (cellular), satellite

25

Define packet switching

Hosts break application-layer messages into packets, forward them from one router to the next across links

26

Briefly describe the store and forward approach

The entire packet must arrive at the router before it can be transmitted on the next link

27

Give the equation for calculating how long it takes to transmit a packet using store-and-forward:
a) one way
b) 2 ways

L = packet size (bits), R = bandwidth (bps)
a) L/R
b) 2*L/R

28

What causes queuing and loss in packet switching?

When arrival rate > transmission rate, packets will queue. If memory buffer fills up, packets can be dropped

29

What happens in circuit switching? Give an example where circuit switching is used

Dedicated rources are allocated to a 'call' between source and destination. There is no sharing of resources and the circuit is idle if not used a call. It is used in telephone networks

30

What is FDM?

Frequency Division Multiplexing - each user occupies a certain band of frequencies

31

What is TDM?

Time Division Multiplexing - time is divided into frames of a fixed duration which is divided into a fixed number of timeslots. Each connection occupies one timeslot

32

Given a 1Mb/s link and each user needing 100kb/s, how many users can be supported in circuit switching?

10 users by pre-allocating

33

Give 2 reasons why packet switching is good for 'bursty data'

1. It allows resource sharing
2. It is simpler, has no call setup

34

State a problem with packet switching

Excessive congestion is possible due to packet delay and loss meaning protocols are needed for rdt and congestion control

35

Which is the provider and which is the customer in the following:
1. Global ISP & Regional ISP
2. Regional ISP & Access ISP

1. Global ISP = provider Regional ISP = customer
2. Regional ISP = provider Access ISP = customer

36

What is tier-1 in an internet structure?

Commerical ISPs e.g. AT&T

37

What is a content provider network?

A private network that connects its data centers to the internet by often bypassing tier-1 and regional ISPs, e.g. Google

38

What is a message in the ___ layer called?
1. Link
2. Network
3. Transport
4. Application

1. Frame
2. Datagram
3. Segment
4. Message

39

Why use layering as an architecture?

When dealing with complex systems, an explicit structure allows identification and relationship between the system's pieces. Modularisation eases maintenance and updating of the system

40

Give formula to calculate nodal processing delay

nodal processing delay = queueing delay + transmission delay (L/R) + propagation delay (length of link/propagation speed) + processing delay

41

Give the formula to calculate traffic intensity

L = packet length (bits), R = link bandwidth (bps), a = average packet arrival rate
= La/R

42

If La/R ~ 0, what does this mean?
Answer the same for if La/R is 1 or > 1?

0 = average queueing delay is small
1 = average queueing delay is large
> 1 = more work arriving than can be serviced

43

What is throughput?

The rate (bps) at which bits are transferred between sender/receiver (either as an average or instantaneous measurement)

44

When comparing capacity, what does R(s) and R(c) represent?

R(s) = capacity of server link (bps)
R(c) = capacity of client link (bps)

45

Give the throughput and bottleneck link for the following:
a) R(s) < R(c)

a) throughput = R(s) (max capacity), bottleneck = R(s)
b) throughput = R(c), bottleneck = R(c)

46

When Rs and Rc for a total of 10 connections are connected to network core (R), how do you calculate the throughput?

Throughput = min(Rs, Rc, R/10)

47

Give two advantages of message segmentation

1. If a single bit error occurs, a whole large message doesn't have to be retransmitted
2. Huge packets would have to be sent into the network and smaller packets have to unfairly queue

48

Give two disadvantages of message segmentation

1. Packets must be put in sequence so more techniques are needed to guarantee this order at destination
2. More packets = more headers which do not change size according to packet size, therefore overall header size across all packets is larger