Revision Library

revisionlibrary.co.uk

GCSE · Computer Science

P1.8 Network Protocols and Layers

OCR J277 · Calculators not allowed · about 95 minutes

Total Marks

Name: _______________________________ Date: ____ / ____ / ______

Answer ALL questions. Show all your working.

A protocol is a set of rules that allows devices on a network to communicate with each other.

(Total for Question 1 is 2 marks)

Different network protocols are used for different purposes.

(a)Which protocol is used to request and receive web pages from a web server?(1)

A) SMTP
B) HTTP
C) FTP
D) IMAP

(b)Which protocol is a secure, encrypted version of the protocol used to transfer web pages?(1)

A) HTTPS
B) FTP
C) TCP
D) POP3

(c)Which protocol is used to send email from a client to a mail server?(1)

A) SMTP
B) HTTP
C) FTP
D) DNS

(d)Which protocol is used to transfer files between computers on a network?(1)

A) FTP
B) SMTP
C) IMAP
D) HTTPS

(Total for Question 2 is 4 marks)

State the purpose of the IMAP and POP3 protocols.

(Total for Question 3 is 3 marks)

Explain why HTTPS is used instead of HTTP when a customer enters their card details on an online shop's website.

(Total for Question 4 is 3 marks)

Networks are often described using a layered model, such as the four-layer TCP/IP model, rather than as one single set of rules.

(a)Name the four layers of the TCP/IP model, in order, starting with the layer closest to the user's application.(4)

(b)State one benefit of organising network communication into separate layers like this, rather than as one single set of rules.(2)

(Total for Question 5 is 6 marks)

Different protocols operate at different layers of the TCP/IP model.

(a)State one protocol that operates at the application layer, and state its purpose.(2)

(b)State the protocol that operates at the transport layer to break data into segments, add sequence numbers, and check that all segments arrive correctly.(1)

(c)State the protocol that operates at the internet layer to address and route packets between networks.(1)

(Total for Question 6 is 4 marks)

TCP and UDP are both transport-layer protocols, but they work differently.

(Total for Question 7 is 4 marks)

A school is choosing a protocol to stream a live assembly to parents watching at home in real time.

(Total for Question 8 is 3 marks)

A file of 2400 bytes is split into equal-sized packets for transmission across a network. Each packet has a fixed overhead of 40 bytes of header data, in addition to its share of the file's payload data.

(a)The file is split into 6 equal packets of payload data. Calculate the number of payload bytes carried in each packet.(2)

(b)Calculate the total size of one packet, including its header, and hence calculate the percentage of one packet's total size that is header data. Give your answer to 1 decimal place.(3)

(Total for Question 9 is 5 marks)

When a large file is sent across the internet using packet switching, it is broken into many separate packets before being sent.

(Total for Question 10 is 5 marks)

Every device connecting to the internet needs both a MAC address and an IP address.

(Total for Question 11 is 4 marks)

Explain the purpose of DNS (the Domain Name System), and why it is needed.

(Total for Question 12 is 3 marks)

A programmer is designing a new online multiplayer racing game. Player position updates must be sent between players many times per second; a slightly late or missing update simply causes a brief visual glitch, and a newer update quickly replaces it. State, with a reason, whether TCP or UDP should be used for sending these position updates.

(Total for Question 13 is 3 marks)

Describe, step by step, what happens at each layer of the TCP/IP model when a user's laptop sends an email to a friend on a different network, from the moment the user clicks 'Send' to the moment the data leaves the laptop.

(Total for Question 14 is 6 marks)

A librarian downloads a 45 megabyte software update over a network connection with an average transfer rate of 15 megabits per second.

(Total for Question 15 is 4 marks)

A protocol is a set of agreed rules, but standards also matter for hardware. State one reason why network hardware, such as network cards and routers, from different manufacturers needs to follow common standards as well as common protocols.

(Total for Question 16 is 2 marks)

State what the term 'bandwidth' means, and state the unit it is normally measured in for a home internet connection.

(Total for Question 17 is 2 marks)

Describe the client-server request-response cycle that takes place when a browser loads a web page, using the terms request and response in your answer.

(Total for Question 18 is 3 marks)

A router forwards packets between different networks. Explain why a router needs to examine the IP address in the header of each packet it receives.

(Total for Question 19 is 3 marks)

State two items of information typically included in a TCP segment's header, other than the payload (data) itself.

(Total for Question 20 is 2 marks)

A TCP segment's header is 20 bytes; a UDP header is only 8 bytes. A live voice call app sends 1000 packets, each carrying 160 bytes of audio payload data.

(a)Calculate the total header overhead, in bytes, saved across these 1000 packets if UDP is used instead of TCP.(3)

(b)Explain why this saving in overhead is particularly useful for a live voice call.(1)

(Total for Question 21 is 4 marks)

Mark scheme · P1.8 Network Protocols and Layers

Question 1

B1 a protocol is an agreed set of rules/standards, oe
B1 that governs how data is formatted, sent and received between devices, so that devices made by different manufacturers can understand each other, oe
Answer: A protocol is an agreed set of rules that governs how data is formatted, sent and received, so devices from different manufacturers can communicate.

Question 2

(a) B1 B cao
(a) Answer: B) HTTP
(b) B1 A cao
(b) Answer: A) HTTPS
(c) B1 A cao
(c) Answer: A) SMTP
(d) B1 A cao
(d) Answer: A) FTP

Question 3

B1 IMAP and POP3 are both protocols used to retrieve/receive email from a mail server, oe
B1 IMAP keeps a synchronised copy of email on the server, so the same mailbox can be viewed from multiple devices, oe
B1 POP3 typically downloads email from the server onto one device (and may then remove it from the server), oe
Answer: Both retrieve email from a mail server; IMAP keeps mail synchronised on the server for viewing on multiple devices, while POP3 downloads mail onto a single device.

Question 4

B1 HTTPS encrypts the data sent between the browser and the web server, oe
B1 so that if the data is intercepted while travelling across the network, it cannot be read/understood by the attacker, oe
B1 this protects sensitive data, such as card details, from being stolen, whereas plain HTTP sends data unencrypted, oe
Answer: HTTPS encrypts data sent between the browser and server, so intercepted card details cannot be read, unlike plain HTTP which sends data unencrypted.

Question 5

(a) B1 Application layer
(a) B1 Transport layer
(a) B1 Internet (network) layer
(a) B1 Link (network interface) layer
(a) Answer: Application, Transport, Internet, Link (in that order).
(b) B1 each layer can be developed, updated or replaced independently, as long as it still uses standard interfaces to communicate with the layers above and below it, oe
(b) B1 this makes it easier for hardware and software from different manufacturers to work together, and simplifies designing/troubleshooting network software, oe
(b) Answer: Layers can be developed and updated independently, using standard interfaces between them, which makes it easier for different manufacturers' hardware/software to work together.

Question 6

(a) B1 a valid application-layer protocol, e.g. HTTP, HTTPS, FTP, SMTP, IMAP, POP3
(a) B1 a correct purpose for the protocol named, matching its standard use
(a) Answer: For example, HTTP - used to request and transfer web pages between a browser and a web server.
(b) B1 TCP, cao
(b) Answer: TCP
(c) B1 IP, cao
(c) Answer: IP

Question 7

B1 TCP is connection-oriented: it establishes a connection, numbers segments, and gets acknowledgements, resending any segment that is lost, so data always arrives complete and in order, oe
B1 this makes TCP reliable but adds overhead/delay, so it is used where accuracy matters more than speed, e.g. loading a web page or sending a file, oe
B1 UDP does not establish a connection, number segments or resend lost data, so some data may arrive out of order or be lost, oe
B1 this makes UDP faster with less overhead, so it is used where speed matters more than perfect accuracy, e.g. live video/voice calls or online gaming, oe
Answer: TCP is connection-oriented and reliable (acknowledges and resends lost data) but slower; UDP has no connection or resending, so it is faster but less reliable, suiting real-time applications like video calls or gaming.

Question 8

B1 UDP, cao
B1 because it has less overhead/does not wait for acknowledgements or resend lost data, so it transmits data faster, oe
B1 for a live stream, keeping the video/audio smooth and close to real time matters more than every single packet arriving perfectly (a brief glitch is preferable to a delay), oe
Answer: UDP, because it transmits with less delay than TCP, and for a live stream staying close to real time matters more than every packet arriving perfectly.

Question 9

(a) M1 2400 / 6 shown as the method, oe
(a) A1 400, cao
(a) Answer: 400 bytes
(b) M1 total packet size = 400 + 40 = 440 bytes, oe
(b) M1 (40 / 440) x 100 shown as the method, oe
(b) A1 9.1%, awrt
(b) Answer: 440 bytes total; 9.1% is header data

Question 10

B1 each packet is given a header containing information such as the sender's and receiver's IP addresses and a sequence number, oe
B1 packets from the same file may travel across the network by different routes, depending on which path is fastest/least congested at that moment, oe
B1 because of this, packets can arrive at the destination out of order, or later than packets sent after them, oe
B1 the sequence number in each packet's header allows the receiving device to reassemble the packets into the correct original order, oe
B1 if a packet does not arrive/is lost, the receiving device can request that it be resent (using a reliable protocol such as TCP), oe
Answer: Packets carry header data including sequence numbers and IP addresses, may travel by different routes and arrive out of order, and are reassembled correctly at the destination using their sequence numbers; missing packets can be requested again.

Question 11

B1 a MAC address is a unique identifier assigned to a network interface at the point of manufacture, and it never changes, oe
B1 an IP address identifies a device's location on a particular network, and can change (e.g. each time the device joins a different network), oe
B1 a MAC address alone cannot be used to route data across the internet, because it does not indicate which network the device is on, oe
B1 so the IP address is used to route data between networks, while the MAC address is used to deliver the data to the correct device once it reaches the destination network, oe
Answer: A MAC address is a permanent, manufacturer-assigned identifier used to deliver data to the right device on a network; an IP address identifies the device's network location and can change, so it is used to route data across the internet.

Question 12

B1 DNS translates/converts a human-readable domain name (e.g. www.example.co.uk) into the numeric IP address of the web server hosting that site, oe
B1 it works like a directory/look-up service that the browser queries before it can connect to the correct server, oe
B1 it is needed because domain names are much easier for humans to remember and type than long numeric IP addresses, while computers still route data using IP addresses, oe
Answer: DNS translates a human-readable domain name into the numeric IP address of the correct server, acting as a directory service, because domain names are easier for humans to remember than IP addresses.

Question 13

B1 UDP, cao
B1 because it does not wait for acknowledgements or resend lost data, so updates are sent with the lowest possible delay, oe
B1 which suits this scenario, since a missing update is quickly replaced by the next one and does not need to be resent, whereas TCP's extra reliability would only add unnecessary delay, oe
Answer: UDP, because its lower delay suits frequent, quickly-superseded updates, whereas TCP's reliability features would only add unwanted delay here.

Question 14

Level 1 (1-2): One or two layers, or one relevant protocol, are named with little explanation of what happens at each stage. Little relevant technical terminology is used.
Level 2 (3-4): Several layers are named in a broadly sensible order, with some explanation of what happens at each one, but the description may be incomplete or lack detail in places. Relevant technical terminology is generally used correctly.
Level 3 (5-6): A clear, well-ordered description covers all four layers, correctly names at least one relevant protocol at each of the application, transport and internet layers, and explains what each layer adds/does to the data. Relevant technical terminology is used correctly throughout.
Indicative content:
- Application layer: the email software uses a protocol such as SMTP to format the email message ready to be sent
- Transport layer: TCP breaks the message into smaller segments, adds sequence numbers, and will later request that any lost segments are resent
- Internet layer: IP adds the source and destination IP addresses to each segment (now called a packet), so it can be routed across networks
- Link layer: the packets are converted into a form (e.g. as frames with a MAC address) that can be transmitted over the laptop's physical connection, such as Wi-Fi or Ethernet
- The data then leaves the laptop and is passed on, hop by hop, across intermediate devices until it reaches the friend's network
- A strong answer explains that each layer adds its own information/header to the data before passing it down to the next layer

Question 15

M1 45 megabytes correctly converted to megabits: 45 x 8 = 360 megabits, oe
M1 360 / 15 shown as the method, oe
A1 24, cao
B1 correctly stated in seconds, cao
Answer: 24 seconds

Question 16

B1 so that hardware made by different manufacturers can physically connect to, and correctly interpret signals from, the same network, oe
B1 without shared standards, a customer would be limited to buying every piece of networking hardware from a single manufacturer, which would be far more restrictive/expensive, oe
Answer: Common standards let hardware from different manufacturers connect and communicate correctly; without them, customers would be locked into buying all their equipment from one manufacturer.

Question 17

B1 bandwidth is the maximum amount of data that can be transferred over a network connection in a given time, oe
B1 megabits per second (Mbps), or an equivalent unit (e.g. Gbps), oe
Answer: Bandwidth is the maximum amount of data that can be transferred in a given time, usually measured in megabits per second (Mbps).

Question 18

B1 the browser (client) sends a request to the web server, specifying which page/resource it wants, oe
B1 the web server processes the request and sends back a response, which contains the requested page/data (or an error if it cannot be found), oe
B1 the browser then displays the page contained in the response to the user, oe
Answer: The browser (client) sends a request for a page to the web server; the server sends back a response containing the page, which the browser then displays.

Question 19

B1 the destination IP address tells the router which network the packet needs to reach, oe
B1 the router uses this address to decide the best next step/route to forward the packet along, towards its destination, oe
B1 without reading the IP address, the router would not know where to send the packet next, so it could not be delivered to the correct network, oe
Answer: The router reads the destination IP address to decide which network/route to forward the packet along; without it, the router could not deliver the packet to the correct destination.

Question 20

B1 any one of: source port number, destination port number, sequence number, acknowledgement number, checksum
B1 a second, different item from the same list
Answer: For example: sequence number and destination port number.

Question 21

(a) M1 20 - 8 = 12 bytes saved per packet, oe
(a) M1 12 x 1000 shown as the method, oe
(a) A1 12000, cao
(a) Answer: 12000 bytes
(b) B1 less header data to send/process per packet means less delay, helping keep the call smooth/close to real time, oe
(b) Answer: Less header overhead means less data to send and process per packet, reducing delay and helping keep the live call smooth.