1
Q
What is ICANN?
A
Internet Corporation for Assigned Names and Numbers (ICANN) is the administrative entity controlling the allocation of internet addresses, domain names and protocol port numbers.
2
Q
What network protocols are included in TCP/IP protocol suite?
A
ARP: translates IP address to MAC address
IP: routes data packets from one address to another
ICMP: provides low-level support to IP (error message, routing, debugging)
UDP: provides non-verified, one-way data delivery
TCP: provides reliable, full duplex, flow controlled, error corrected data delivery
3
Q
Give the protocols and data unit for application layer.
A
Protocols: HTTP/HTTPS/FTP/SMTP/LDAP/DHCP/DNS
Data unit: message
4
Q
Give the protocols, data unit and address name for transport layer.
A
Protocols: TCP/UDP
Data unit: segment
Address name: port number (2 bytes)
5
Q
Give the protocols, data unit and address name for network layer.
A
Protocols: IP/ICMP
Data unit: packet
Address name: IPv4 (4 bytes), IPv6 (16 bytes)
6
Q
Give the protocols, data unit and address name for data link layer.
A
Protocols: Ethernet/WiFi/PPP/ARP
Data unit: frame
Address name: MAC address (6 bytes)
7
Q
Give the protocols and data unit for physical layer.
A
Protocols: IEEE 802.3
Data unit: bits
8
Q
The acronym OSI stands for
A
Open Systems Interconnection, it is a conceptual framework describing how data moves through the network although it is not implemented in real world.
9
Q
Explain OSI Model.
A
OSI model is a set of protocols that allows any two different systems to communicate regardless of their underlying architecture.
It consists of 7 layers, which are generally divided into two groups- host layers and media layers:
Host layers- Application, Presentation, Session
Media layers- Transport, Network, Data Link, Physical
10
Q
Explain Physical Layer.
A
It is responsible for moving individual bits from one node to the next.
It deals with
- mechanical and electrical specifications of the interface and transmission media.
- bit encoding (electrical/optical)
- transmission rate: bits/sec
- physical topology (mesh/star/ring/bus)
- transmission mode (simplex/half duplex/full duplex)
11
Q
Explain Data Link Layer.
A
It is responsible for transforming the physical layer to a reliable link of data and to make physical layer appear error-free to upper layer (Network layer).
It deals with
- Framing : dividing the stream of bits into manageable data units called frames
- Physical addressing: either add address of sender and receiver to the frame or only sender’s address (if broadcast) or address of gateway router (if receiver is not on sender’s network)
- Flow control: if receiver is slower than sender must reduce its transmission speed
- Error control: if a frame is damaged/lost, ask for retransmission. If a frame is duplicated, drop it.
- Access control: if two or more devices are connected to the link, the Data Link Layer protocols decide which device has control over the link
12
Q
Explain Network Layer.
A
It is responsible for source to destination delivery of data packets, possibly across multiple links.
It deals with
- Logical addressing: if the receiver and sender are on two different networks
- Routing: sending data packets to different networks through network-connecting devices (routers/switches) by choosing best path
13
Q
Explain Transport Layer.
A
It is responsible for process-to-process delivery, making sure that the message is intact and in order.
It deals with
- Sending correct port address
- Segmentation and orderly reassembly of message
- Connection control: connectionless/connection-oriented
- Flow control: if receiver is slower than sender’s transmission speed is managed to match receiver’s receiving speed
- Error control: if a segment is damaged/lost, ask for retransmission.
14
Q
Explain Session Layer.
A
It is responsible for establishing, managing and terminating the sessions.
It deals with
- Synchronization
- Interhost communication
- Maintaining session
15
Q
What are the checkpoints added in Session layer?
A
If a system is sending a file of 2000 pages, then instead of sending all of it in one batch and checking for error afterwards, a checkpoint can be added after every 100 pages so that error checking is performed after every 100 pages.
If the error is detected, only those 100 pages are retransmitted instead of entire file.
16
Q
Explain Presentation Layer.
A
It is responsible for the syntax and semantics of the data.
It deals with
- Encryption and decryption
- Compression
- Data presentation
17
Q
Explain Application Layer.
A
It is responsible for providing the user interface and services (such as email) to the end user.
It deals with various services such as email, file transfer etc
18
Q
Explain how process ‘a’ on client A connects to process ‘j’ on client J.
A
- The application layer at client A sends the data down to the transport layer which then encapsulates the message with the source and destination port addresses (a and j). A TCP packet is formed at this point.
- Transport layer sends the packet to network layer which encapsulates it in logical addresses (A and J). An IP datagram is formed here.
- Network layer sends the packet to data link layer which encapsulates it in physical (MAC) addresses (20 and 55, lets say). The MAC addresses are generated by consulting ARP table. A frame is formed here.
- Data link layer sends the frame to physical layer which streams the bits to all the routers on its link and the router whose physical address is 55 keeps the packet, all other discard it. If the client J is on same link then router will directly send the packet to client, otherwise will forward it to the next router by changing source and destination address. Source address will be the sending router’s address and destination address will be next router’s address.
- Once we reach the final correct router, the frame is discarded by network layer (getting rid of physical addresses) and logical address is checked. The packet is forwarded to node whose logical address is J.
- When the package reaches J’s transport layer, logical address is discarded and port address is checked. The data is then sent to the port j.
19
Q
Which layer provides the route determination?
A
Network layer
20
Q
Which layer provides the flow control?
A
Both data link and transport layers
21
Q
Which layer provides the interface to transmission media?
A
Physical layer
22
Q
Which layer provides the access for end user?
A
Application layer
23
Q
Which layer provides the reliable process-to-process message delivery?
A
Transport layer
24
Q
Which layer provides the route selection?
A
Network layer
25
Which layer provides the frame definition?
Data link layer
26
Which layer provides the user services such as email and file transfer?
Application layer
27
Which layer provides the transmission of bit streams across physical medium?
Physical layer
28
Which layer provides the direct communication with user's application program?
Application layer
29
Which layer provides the error correction and retransmission?
Transport layer
30
Which layer provides the mechanical, electrical and functional interface?
Physical layer
31
Which layer provides the responsibility for carrying frames between adjacent nodes?
Data link layer
32
Which layer in OSI model provides the formatting and code conversion services?
Presentation layer
33
Which layer in OSI model provides the establishment, management and termination of services?
Session layer
34
Which layer in OSI model provides the reliable transmission of data?
Transport layer
35
Which layer in OSI model provides the log-in and log-out procedures?
Session layer
36
Which layer in OSI model provides the independence from differences in data representation?
Presentation layer
37
Which layer divides itself into two sublayers? Name the sublayers.
Data link layer divides itself into two sublayers:
Logical Link Control (LLC) and
Media Access Control (MAC)
38
A 100-byte message is sent using TCP/IP protocol suite. If the protocol adds a 10-byte header at each layer, what is the efficiency of the system (ratio of number of useful bytes to number of total bytes)?
Message is of 100 bytes: number of useful bytes are 100.
```
Transport layer adds 10 bytes,
Network layer adds 10 bytes,
Data link layer adds 10 bytes
Total header= 30 bytes
Total bytes = 100+30=130
```
Efficiency= (100/130) * 100 = 77%
39
What difference does it make to the network layer if the underlying data link layer provides a connection-oriented service or connectionless service?
If the DLL provides a connection-oriented service then the network layer must precede the information transfer with a connection setup procedure.
On the other hand, if the connectionless service is provided, then each data transfer is an independent request and network layer cannot assume the correctness of data sent.
40
If the transmission channels become virtually error-free, do we still need the data link layer?
Yes, as DLL provides the framing of data and flow control over a transmission channel.
Also, in a multiple access medium (LAN), DLL is required to coordinate the access to the shared medium among multiple users.
41
Does ethernet provide connectionless or connection-oriented services?
Ethernet provides connectionless service.
42
If all computer in a city are to be communicated using radio transmissions froms single HF antenna tower. Which layer is more appropriate for this situation- DLL or Network?
The DLL is concerned with data transfer across single hop whereas network layer is concerned with data transfer across a network in multiple hops per path in general.
As the connection from radio antenna is direct, DLL is more appropriate.
43
Why is it useful for processes such as email and file transfer to have 'well-known' port address?
So that transport layer does not have to use a new protocol just to find out the port number
44
If a computer is moved from inside of one office to another office, which address is bound to change- physical or IP?
Only IP address will change as physical address is globally unique to the computer's NIC.
45
If all computer in a city are to be communicated using large number of small antennas using radio transmission. Which layer is more appropriate for this situation- DLL or Network?
Network layer would be appropriate as it provides routing of data packets across network to traverse a number of network nodes.
46
What does the Ethernet frame contain?
Ethernet frame contains 7 fields:
```
Preamble (7 bytes)
Start Frame Delimiter SFD (1 byte)
Destination MAC address (6 bytes)
Source MAC address (6 bytes)
Length/ type of PDU (2 bytes)
Payload (min 46 bytes, max 1500 bytes)
CRC (4 bytes)
```
47
Explain the preamble and SFD in Ethernet frame.
Preamble is the first field in Ethernet frame and contains 7 bytes of alternation 0s and 1s to alert the receiver to sync its input timing.
Start Frame Delimiter (SFD) is second field in Ethernet frame and is of 1 fixed byte: 10101011. It warns the receiver that this is the last chance to sync the input timing and last 2 bits (11) alert the receiver that the next field is destination address.
48
What is the minimum and maximum length of Ethernet frame?
As preamble and SFD are Physical layer headers, we have
- DA and SA (each 6 bytes),
- length of PDU (2 bytes)
- CRC (4 bytes)
total 18 bytes of header.
As minimum and maximum payload is 46 and 1500 bytes, minimum and maximum length of Ethernet frame is 64 and 1518 bytes.
49
Explain MAC address.
The physical address of a network adapter card (burnt in the ROM chip of card) is called MAC address. It is 6 bytes long and represented in hexadecimal format.
Eg. 00:21:70:6F:06:F2
The first 3 bytes are assigned to the manufacturer by IEEE and called OUI (Organizationally Unique Identifier).
The last 3 bytes are assigned by manufacturer to chip sequentially, which allows for uniqueness. This gives 2^24 unique addresses (~16.7 mil unique addresses per organization).
The MAC addresses are physical and cannot be changed, they belong to the device (As opposed to IP addresses). They are used for local LAN communication.
50
How to recognize if an address is unicast/multicast/broadcast from MAC address?
A source address is always unicast - the frames come from only one node.
A destination address can be unicast or multicast; unicast destination address defining only one recipient whereas multicast destination address defines multiple recipients.
LSB of the first byte of MAC is 0 --> unicast
LSB of the first byte of MAC is 1 --> multicast
All bits in MAC are 1 --> broadcast
51
# Define the type of address:
(i) 4A:30:10:21:10:1A
(ii) 47:20:1B:2E:08:EE
(iii) FF:FF:FF:FF:FF:FF
(i) The first byte is 4A which in binary is 0100 1010 and the LSB is 0, hence this is unicast address
(ii) The first byte is 47 which in binary is 0100 0111 and the LSB is 1, hence this is multicast address
(iii) All the bits are 1s hence this is broadcast address
52
How is the address 47:20:1B:2E:08:EE sent out on a line?
The address transmission is left to right and for each byte, LSB is sent first so that the receiver knows if its a unicast/multicast address.
47: 0100 0111
20: 0010 0000
1B: 0001 1011
2E: 0010 1110
08: 0000 1000
EE: 1110 1110
is sent as:
```
1110 0010
0000 0100
1101 1000
0111 0100
0001 0000
0111 0111
```
53
Explain CSMA/CD.
Carrier Sense Multiple Access/Collision Detection (CSMA/CD) is the access method for traditional Ethernet which allows stations to share a channel and also eliminate data collisions.
Multiple stations on Ethernet are connected physically by either bus/star topology and logically by bus topology, hence the medium is shared between stations and only one station can use the medium at a time.
All stations receive the frame but only real destination keeps it while others drop it.
The chance of collision is reduced if a station senses the medium before trying to use it. If a station finds the medium idle and sends the data frame, it checks if the transmission was successful. If so, the station is done otherwise the frame is retransmitted.
54
How do stations on a WLAN defer sending their data if one station acquires access?
When a station sends a Request To Send (RTS) frame, it includes the duration of time that it needs to occupy the channel.
The stations that are affected by this transmission create a timer called Network Allocation Vector (NAV) that shows how much time must pass before the stations are allowed to check the channel for idleness.
Each time a station access a system and sends an RTS frame, other stations start their NAVs. Consequently, each station, before sensing the medium to see if its idle, first checks its own NAV to see if its expired.
55
What is the hidden station problem and how it is solved?
If there are two stations B and C whose transmission ranges are spherical in shape and they are situated such that they are both out of each others' ranges but station A is in the area covered by both B and C's transmission range, then B and C are called hidden stations from each other with respect to A.
When B is in the middle of tranmitting data to A and C sends its RTS to A (as it thinks medium is free), there is possibility of collision.
The solution is to use handshake frames (RTS and CTS). Before B can send the data, B sends Request To Send (RTS) to A and A in return sends Clear to Send (CTS) to B which contains duration of data transmission from B to A.
As the CTS from A also reaches C, station C knows that some hidden station is using the channel and refrains from data transmission until CTS duration is over.
56
What is exposed station problem?
If there are two stations B and C whose transmission ranges are spherical in shape and they are situated such that a station A is in reach of both but station D is within station C's range, the station C is exposed to transmission from A to B.
When station A is transmitting to B and station C has some data for D, station C hears A's RTS but as A starts sending data, C cannot hear D's CTS because of the collision. It remains exposed until A finishes sending its data.
57
Explain the two types of networks used by Bluetooth.
Bluetooth uses piconet and scatternet.
Piconet is an ad hoc network, created spontaneously, and can have upto 8 stations. Only one of them is primary and rest are secondaries, which sync their clocks and hopping sequences with primary. Apart from 7 secondaries, 8 more stations can be introduced in the parked state (they are synced with primary but cannot participate in conversation). If an active secondary goes to parked state, a station from parked state can become secondary.
Scattenet is made when piconets are combined. Secondary in one piconet can be primary in another piconet. It can receive messages from primary of first piconet and deliver it to secondaries of second piconet.
58
Explain the ARP protocol.
The Address Resolution Protocol (ARP) is level 2 protocol (data link layer) and maps the logical IP address to physical MAC address and stores this pairing in the ARP table on PCs and routers.
When a host's ARP table is empty or doesn't have the listing for given IP address, it broadcasts the destination IP address on its subnet. If the destination IP address is on same subnet, it responds back with its MAC address, which host stores in its ARP table.
If the destination IP address is not on the same subnet, the default gateway forwards the packet.
59
Explain DNS.
Domain Name System (DNS) is used to resolve domain names into its associated IP address. It is level 5 protocol and can use TCP/UDP protocol.
The system takes the domain name and forwards it to DNS server which returns the IP address.
60
What are the two types of network architecture?
Client-server and peer-to-peer
61
Explain client-server network architecture.
Network is composed of clients, who request for resources, and servers, who provide resources.
This architecture introduces centralized user accounts and more efficient user access control. On the other hand, server failure can render the entire network unusable. It also needs dedicated hardware and software along with expert staff, which adds to the cost.
62
Explain peer-to-peer network architecture.
Network is composed of computers which are peers of each other and there are no dedicated servers. Any individual node can act as a client or server sharing resource.
This architecture is easy to install and incurs low costs than client-server as there is no hardware and software setup and no need of administrators. On the other hand, the network security applies only to the single resource at a time and each machine needs to be backed up individually to protect the data.
63
What are the types of network?
Local Area Network (LAN)- small network covering a single room/ a building/ a group of buildings
Campus Area Network (CAN)- interconnected LANs in a limited geographical area such as a corporation/ government agency/ university campus
Metropolitan Area Network (MAN)- network that connects the users with computer resources in a city
Wide Area Network (WAN)- network that extends over multiple cities and/or countries
64
What are the two types of network protocols?
Physical protocols- describe the medium, connections and signal
Logical protocols- software controlling how and when data is sent and received.
65
How can you see the MAC address of your computer's networking adapters?
Go to command prompt on Windows and type
ipconfig /all
In the response, find Wireless LAN configuration section and check the physical address.
66
Explain IPv4 addresses.
Internet Protocol version 4 is the network-layer protocol and often called logical address. It is 4 bytes long and is written in dotted decimal format.
Eg. 192.168.240.3
These are for distant WAN communication and are also called network addresses as they belong to the network, not to the device (As opposed to the MAC address).
67
Explain simplex vs half duplex vs full duplex communication.
Simplex communications is where we can either send data or receive data- like a radio station. It can only send data, not receive any.
When simplex communication is reversible, i.e., the device which can send data can also be used to receive data, just not at the same time, is called half duplex communication. For eg, Walkie-talkie.
Full duplex communication is where you can send and receive data at the same time- like telephone.
68
Explain Ethernet.
Ethernet refers to the family of protocols that control how data is transmitted over LAN and involves speeds, cables, connectors, equipments. It deals with the physical aspects of the network.
Modern Ethernet uses twisted pair/fiber cable and CSMA/CD logical topology.
Ethernet uses "xx Base T" naming convention where
xx: speed in Mbps
Base: baseband communication (single frequency)
T: type of cable
For eg: 10BaseT is 10Mbps twisted pair and 1000BaseF is 1Gbps fiber optic.
69
Explain network topologies.
Network topologies define the structure and/or properties of a network and can be either physical or logical.
Physical topologies define the physical design of a network like network devices, location, cables etc.
Logical topologies define how data moves through a network like CSMA/CA, CSMA/CD, Ethernet.
70
Explain bus topology.
Its a physical topology where all devices are connected to a single network cable. This single network cable is called bus.
The bus cable requires terminators on the both ends, without which the communication is disrupted. A single point of failure can bring the entire network down.
71
Explain ring topology.
Its a physical topology where all devices are connected to its neighbors in a circular ring fashion.
Each computer is directly connected to 2 other computers and data travels from node-to-node with each computer handling data, either unidirectional or bidirectional.
Failure in a node can bring down the entire network if the ring is unidirectional.
72
Explain star topology.
Its a physical topology in which all devices are connected to a central networking device, usually a switch or hub.
It is used in most small and large networks but the failure in central networking device can bring down the network. Failure in single node affects the node itself, not the enitre network.
73
Explain mesh topology.
Its a physical topology in which all devices are connected to each other by separate cabling.
Although it has high fault-tolerance, it is also expensive to install in LANs and is highly redundant. This topology is used in WANs so that connectivity stays put even during cabling/node failure as all nodes are connected in multiple ways.
74
Explain NIC.
Network Interface Card (NIC) is the network adapter installed on a network device. It is the hardware component that connects a host device to a network.
Anything and everything that connects to a network from a smartphone, tablet, computer, laptop to a server, firewall, router, switch has to have at least one NIC.
NIC provides the physical and electrical, light or radio frequency connections to the network media.
75
Explain hubs.
Hubs are the network devices used to connect other devices together within a network. Like in a star topology, a central hub connects all the host computers with each other through itself. It is OSI layer 1 device.
Hubs were used in early networks and are now replaced by switches. The reason behing this is that hub is a multi-port repeater- if a node A wants to send data to node B, the central hub will flood the network with data from A. This makes hubs less efficient and less secure than switches.
76
Explain Switch.
Switch is the connecting device in a network, just like a hub. It stores the MAC addresses of all the nodes it is connected to in its MAC address table. It is OSI layer 2 device.
When data from node A is to be sent to node B, switch pays attention to the source and destination MAC addresses and sends the data only to node B. This effectively breaks up collision domains.
77
Explain WAP.
Wireless Access Point (WAP) is a bridge that connects wireless devices to a wired network.
In an office where desktop computers are using wired connections by directly connecting to the router, WAPs can be connected to the router and laptops and other wireless devices can then be brought inside network.
WAPs can only provide wireless connections to the host devices, whereas wifi routers can provide wireless as well as wired connections as they have ethernet ports for this purpose.
WAPs do not have firewalls nor do they have DHCP services offering IP addresses to connected devices. Rather, the main router to which WAP is connected to provides IP addresses.
Wifi routers have a WAN port which connects to a modem and provides wireless internet. WAP doesn't have a WAN port but has an Ethernet port and hence cannot connect to modem directly, it must go through a router.
78
Explain Router.
Router is a networking device used to connect different networks together. They route traffic between networks using IP addresses.
Routers use routing protocols to find the best way to get a packet of information from source to destination.
It is OSI layer 3 device.
79
Explain basic firewall.
Firewall protects the LAN from malicious activity on the network by preventing unwanted network traffic to access the LAN. It essentially controls the flow in and out of a network.
Network access rules can be customized by administrator of the network.
A firewall can be a standalone network device or a software on a computer system (network based- protecting entire network or host based- protecting the host machine only).
80
Explain DHCP.
Dynamic Host Configuration Protocol (DHCP) is used by DHCP server to
- dynamically assign IP addresses to hosts connected to a network.
- provide updated DNS server/ gateway/router information
The alternative is static IP.
81
What are the three types of network cabling widely used?
Coaxial
Twisted pair
Fiber optic
82
Explain twisted pair copper cabling.
It contains 4 twisted pairs of wires with RJ-45 connector which looks similar to a telephone jack but is slightly wider.
One of the wires in a pair is positive and another one is negative. They are twisted so as to reduce the electromagnetic interference.
They can be used upto 100 metre maximum distance post which they need some sort of signla repeater.
There are two types of twisted pair cables-
- Unshielded Twisted Pair (UTP): more susceptible to EMI
- Shielded Twisted Pair (STP): less susceptible to EMI
In these types, they have multiple categories with naming convention CAT followed by category number, followed by letter sometimes, like CAT 3, CAT 5e, CAT 6a etc.
Higher the category number, higher the speed and higher the twists to the pair of cable to reduce crosstalk.
83
Explain fiber optic cabling.
These are the glass of plastic fiber that carry signal in the form of light, which makes them have high bandwidth over long distances. Also the signal being optical, these cables are immune from losses due to EMI.
There are two types of cables-
- Multi mode fiber (MMF), which cover shorter distances and carry multiple signals
- Single-mode fiber (SMF), which cover longer distances such as across town, carrying single signal and are much more expensive.
Google Prep
flashcards
Decks in class (6)
# Cards
