Communication & networking (TCP/IP, subnetting, client-server) – A-level Computer Science exam questions with mark scheme

1Core3 marks

The TCP/IP stack has four layers: application, transport, network and link.

State the layer responsible for each of the following. [3]

(i) Routing packets between networks using IP addresses.
(ii) Splitting data into packets and using port numbers to identify the receiving program.
(iii) Defining the format of an HTTP request.

(i) Network layer (allow internet layer)(1)
(ii) Transport layer(1)
(iii) Application layer(1)
Answer: (i) network (internet) (ii) transport (iii) application
Examiner tip: Ports belong to the transport layer; IP addresses belong to the network (internet) layer.

2Standard4 marks

A computer has the IP address 192.168.14.77 and the subnet mask 255.255.255.224.

(a) Calculate the network identifier (network address) of the subnet that this computer is on. Show your working. [2]

(b) The computer needs to send a packet to the IP address 192.168.14.101. Explain how the computer uses the subnet mask to decide whether it can send the packet directly to the destination or must send it to the router. [2]

(a) Working: last octet 77 = 01001101 ANDed with 224 = 11100000 (bitwise AND of the address with the mask) (allow equivalent working, e.g. block size 256 − 224 = 32 and 77 lies in the block 64–95)(1)
(a) 192.168.14.64 (ignore /27 suffix)(1)
(b) The subnet mask is ANDed with the destination IP address (and with its own address) to find the network identifier of each (allow: the mask is applied to / used to pick out the network part of the addresses)(1)
(b) If the network identifiers match (both on the same subnet) the packet is sent directly; otherwise it is sent to the router / default gateway. (For this destination: 101 AND 224 = 96, so 192.168.14.96 ≠ 192.168.14.64; it must go via the router. Award the mark for the general rule or for the correct conclusion with reason.)(1)
Answer: (a) 192.168.14.64 (b) AND the destination address with the mask to get its network ID (192.168.14.96); it differs from 192.168.14.64, so the packet is sent to the router
Examiner tip: Show the AND in binary for the working mark; just writing the answer risks losing a mark.

3Standard4 marks

A school network can be set up as client-server or peer-to-peer.

(a) Give two advantages of using a client-server network rather than a peer-to-peer network. [2]

(b) Give two disadvantages of using a client-server network compared with a peer-to-peer network. [2]

(a) Any one advantage from: security / access rights managed centrally; user accounts managed centrally so users can log in from any client; files stored centrally so backups are done centrally; software installs and updates managed centrally; servers are dedicated / more powerful so services are more reliable(1)
(a) A second, different advantage from the list above (do not credit two points that both just say 'central management' without saying what is managed)(1)
(b) Any one disadvantage from: servers are expensive to buy; specialist staff (network manager) needed to run and maintain them; if the server fails, clients lose access to services (single point of failure); server can become a bottleneck when many clients make requests(1)
(b) A second, different disadvantage from the list above(1)
Answer: Advantages (any two): central security, central backup, central user accounts, central software updates. Disadvantages (any two): cost of server, specialist staff needed, single point of failure, server bottleneck
Examiner tip: Say what is managed centrally (backups, security, user accounts); "it is easier" or "it is slow" is too vague to score.

4Hard6 marks

A router connects four subnets of a company network (see diagram). Every host on the network uses the subnet mask 255.255.255.192.

Four hosts have these IP addresses:

Host A: 192.168.10.37
Host B: 192.168.10.70
Host C: 192.168.10.120
Host D: 192.168.10.200

(a) Write the last octet of the subnet mask in binary and state how many bits of the IP address form the network identifier. [2]

(b) Calculate the network identifier of Host B. Show your working. [2]

(a) 11000000(1)
(a) 26 bits (8 + 8 + 8 + 2)(1)
(b) Working: 70 = 01000110 ANDed with 11000000 (allow equivalent block working: blocks of 64, 70 lies in 64–127)(1)
(b) 192.168.10.64(1)
(c) Host B and Host C(1)
(c) Justification: both B and C have network identifier 192.168.10.64 (120 AND 192 = 64) (allow: both lie in the block 64–127). Stating that A is on 192.168.10.0 and D is on 192.168.10.192 is not required. Do not award the justification mark if the hosts identified are wrong.(1)
Answer: (a) 11000000; 26 bits (b) 192.168.10.64 (c) B and C, both have network ID 192.168.10.64
Examiner tip: Hosts are on the same subnet only if ANDing each address with the mask gives the same network identifier; addresses that 'look close' are not enough.

5Standard5 marks

A user types www.example.com into a web browser.

(a) Describe how the Domain Name System (DNS) is used so that the browser can load the page. [3]

(b) Explain why the DNS is used rather than requiring users to type IP addresses. [1]

(a) The domain name is sent (by the browser / computer) to a DNS server (allow: local cache checked first, then the DNS server queried if not found)(1)
(a) The DNS server looks up the matching IP address (or asks other DNS servers if it does not hold it) and returns the IP address(1)
(a) The browser uses the IP address to contact the web server and request the page(1)
(b) Domain names are easier for humans to remember than IP addresses (allow: the IP address of a server can change without users needing to know)(1)
(c) A port number reserved / assigned for a particular (standard) server-side application or protocol, so clients know which port to connect to (e.g. 80 for HTTP, 443 for HTTPS). Do not require the range 0–1023 or an example; reject 'a port that is commonly used' with no idea of it being reserved for a service(1)
Answer: (a) name sent to DNS server; IP address looked up and returned; browser uses IP address to request page from web server (b) names are easier to remember (c) a port number reserved/assigned to a standard server application protocol, e.g. 443 for HTTPS
Examiner tip: Describe the lookup as name to IP address; the DNS does not fetch the web page.

6Hard5 marks

A company uses a packet-filtering firewall and a proxy server.

(a) Describe how a packet-filtering firewall decides whether to allow a packet. [3]

(b) Explain two ways in which a proxy server can improve the security or performance of the network. [2]

(a) It examines header information of each packet, e.g. source / destination IP address, port number, protocol (at least one field needed)(1)
(a) It compares these with a set of rules (an access control list) set by the network administrator(1)
(a) The packet is allowed through if a rule permits it, otherwise it is dropped / rejected / blocked (ignore references to stateful inspection: not required and does not replace this point)(1)
(b) Any one way, with its benefit, from: hides internal IP addresses (requests appear to come from the proxy) so external attackers cannot directly target internal machines; filters / blocks access to unwanted or malicious websites or content; logs / monitors requests so misuse can be traced; caches frequently requested pages so they load faster and use less Internet bandwidth (allow any other valid way, e.g. all traffic must pass through one point that can be checked)(1)
(b) A second, different way, with its benefit, from the list above(1)
Answer: (a) header checked against rules, accept or drop (b) hides internal addresses/filters content; caching
Examiner tip: Say which header fields are checked; "checks for viruses" describes antivirus, not packet filtering.

Still losing marks on this topic?

Book a free introductory session and we'll find exactly which step is costing you.

Book a free intro