Loading…

Worked solutions

IB Computer Science HL · Paper 3

Computer Science, Higher Level

Session: May 2019
Marks: 30
Duration: 1 hour
Calculator: Not allowed
Questions: 4

Keep your copy of the paper open alongside. We don't reproduce the questions: each part below is labelled to match the May 2019 Global paper, which your school or the IB can provide.

Practice mode

Hide our working until you've had a go.

Questions

Questions 1–4 · 30 marks

Question 1

4 marks

Data tier functions in multi-tier architectures and TCP/IP socket fundamentals

Databases
Networks

(a)

2 marks

Identify two core functions executed by the Data Tier in the EMIS multi-tier architecture:
1) Persistent data storage and transactional retrieval: The data tier executes Data Manipulation Language (DML/SQL) commands to store, query, update, and index emergency incident records and geospatial datasets across relational and spatial database management systems.
2) Enforcing ACID properties and data abstraction: It isolates backend database schemas from application business logic, managing connection pooling, transactional integrity, concurrent locking, and data consistency during simultaneous incident reports.

Mentor tip

Focus on backend data operations: CRUD queries, SQL execution, connection abstraction, and maintaining ACID transaction consistency.

(b)

2 marks

Identify two essential characteristics of a TCP/IP network socket:
1) Endpoint binding to IP address and port: A socket represents an operating system software abstraction uniquely defined by the concatenation of a network IP address and a specific transport-layer port number (e.g., 192.168.1.10:8080).
2) Bidirectional connection-oriented communication: It establishes a full-duplex communication channel between two networked nodes, operating across OSI layers 4 (Transport) and 5 (Session) to stream ordered, acknowledged data packets.

Mentor tip

Highlight the endpoint pairing of an IP address with a port number and its role as a bidirectional transport conduit.

Question 2

8 marks

Software procurement tradeoffs and stateful HTTP connections via URL rewriting

System fundamentals
Web science

(a)(i)

2 marks

Explain one non-cost advantage of purchasing an off-the-shelf emergency response system over a bespoke build:
1) Rapid deployment timeframe: Commercial off-the-shelf software is pre-compiled, tested, and immediately available for installation, enabling Bangbai to deploy emergency capabilities in weeks rather than enduring lengthy multi-year software development cycles where lives could be lost.
2) Proven software reliability: Off-the-shelf packages have undergone rigorous field testing across numerous municipal deployments, offering comprehensive documentation, established user communities, and fewer unforeseen software defects.

Mentor tip

Explain the speed of implementation or extensive real-world testing without mentioning financial costs.

(a)(ii)

2 marks

Explain one non-cost disadvantage of selecting an off-the-shelf emergency response system:
1) Incompatible interfaces with legacy municipal systems: Off-the-shelf platforms are built to generic industry standards and may not integrate with Bangbai's existing bespoke dispatch hardware or specialized emergency sensors without complex integration adapters.
2) Inability to modify closed source code: If Bangbai requires emergency features unique to its local geography (e.g., localized monsoon flood models), municipal engineers cannot alter the underlying codebase without relying on external third-party vendor release roadmaps.

Mentor tip

Focus on rigid closed source code or integration difficulties with existing municipal infrastructure.

(b)

4 marks

Explain two advantages of using URL rewriting to maintain state across stateless HTTP connections:
1) Robust operation independent of client cookie settings: Standard HTTP session persistence relies on browser cookies; however, users frequently disable cookies for privacy reasons or utilize legacy mobile browsers that lack cookie support. URL rewriting embeds the session token directly into the query string parameter, ensuring session continuity across all devices without requiring client-side storage.
2) Enhanced client privacy and forensic auditing: URL rewriting avoids persisting state tokens on the client device filesystem where they could be inspected or harvested by malware. Furthermore, having session tokens explicitly present within server access logs allows network engineers to reconstruct user transaction sequences and trace dropped emergency connections during incident post-mortems.

Mentor tip

Present two distinct advantages: functioning when cookies are disabled or unsupported, and avoiding client-side cookie storage vulnerabilities while enhancing server-side audit logs.

Question 3

6 marks

Technical impediments of VPN tunneling on citizen incident reporting apps

Networks

6 marks

Explain why routing citizen reporting traffic through a VPN reduces the app's operational efficacy:
1) Masking true origin IP addresses: A VPN routes packets through an intermediary external relay server, replacing the citizen's authentic IP address with the exit node IP address. If the smartphone's internal GPS is disabled or indoors, the emergency dispatch server cannot perform network-based geolocation or verify the physical location of the reporter.
2) Latency overhead and connection drops: Encapsulating, tunneling, and encrypting network packets increases protocol overhead and packet header sizes, introducing transmission latency and elevating the probability of dropped connections over volatile cellular links during crises.
3) Port blocking and protocol filtering: Commercial VPN providers often apply restrictive filtering rules or block UDP and non-standard communication ports, inadvertently disabling VoIP streams, video feeds, and real-time push telemetry required for emergency assessment.
4) Obscuring geographic telemetry validation: Automated dispatch engines cross-reference reported GPS coordinates against the caller's IP geolocation to detect fraudulent calls; a mismatch caused by a foreign VPN exit node can delay dispatch triage.
5) Anonymity-induced abuse: Perceived absolute anonymity can encourage malicious actors to submit false reports, consuming emergency dispatch personnel and diverting finite first responders away from authentic life-threatening incidents.
6) Third-party logging liabilities: Unvetted commercial VPN nodes can log sensitive user data and metadata, compromising the confidentiality of whistleblowers reporting dangerous criminal incidents.

Mentor tip

Explain how VPN masking breaks IP-based geolocation, adds encryption overhead/latency, blocks essential streaming ports, and complicates hoax verification.

Question 4

12 marks

Comparative evaluation of server load balancing algorithms for Bangbai EMIS

Networks

12 marks

Evaluate the appropriateness of Client-Side Random versus Source IP Hash load balancing for the Bangbai EMIS citizen reporting server cluster:
1) Operational mechanisms:
- Client-Side Random: The client application retrieves an inventory of available server IP addresses and randomly selects an endpoint for each request without consulting a centralized server-side distributor.
- Source IP Hash: A load balancer computes a mathematical hash from the client's source IP address and destination address, modulo the number of active cluster nodes ( $ServerIndex = Hash (IP_{source}) (mod N)$ ), routing all packets from that IP to a deterministic server.
2) Session persistence and stateful operations:
- Source IP Hash maintains intrinsic session stickiness: subsequent requests from the same user land consistently on the same backend node, preserving active in-memory incident reporting drafts and file uploads without requiring clustered session synchronization.
- Client-Side Random lacks session affinity: consecutive requests may land on different servers, terminating half-completed reporting forms unless expensive shared distributed caching is implemented.
3) Vulnerability to asymmetric traffic and NAT aggregation:
- Source IP Hash suffers from severe load imbalance when hundreds of citizens report from behind a single shared cellular carrier NAT or municipal proxy gateway; all requests share an identical public IP, routing all traffic to a single server and creating bottleneck hotspots.
- Client-Side Random distributes requests evenly across all cluster nodes regardless of client IP concentration, naturally preventing single-server starvation caused by carrier NAT.
4) Security, denial-of-service, and single points of failure:
- Client-Side Random exposes all internal cluster server IP addresses directly to the public client, allowing malicious actors to target specific backend nodes with directed Denial of Service (DoS) floods.
- Source IP Hash operates behind a shielded reverse proxy that conceals backend topologies. However, that centralized reverse proxy represents a single point of failure (SPOF) unless deployed with redundant hardware pairs.
5) Cost, implementation complexity, and synthesis:
- Client-Side Random requires zero expensive hardware load balancers, making it cheap to implement, but its lack of persistence and heightened security vulnerability make it inadequate for high-reliability emergency systems.
- Recommendation: Source IP Hash deployed on an active-passive hardware load balancer pair is far superior for Bangbai EMIS, providing critical session continuity for emergency incident uploads, provided that consistent hashing algorithms are employed to gracefully accommodate cluster node scaling.

Node = Hash (IP_{source}) (mod N)

Mentor tip

Compare both algorithms across session persistence, traffic distribution under carrier NAT, security/DoS attack exposure, single points of failure, and conclude with a justified recommendation.

Our analysis

Where the marks are

Each part sorted by the topic it mostly tests. Networks carries the most marks on this paper.

Networks20 marks · 67%
Q1, Q3, Q4
System fundamentals4 marks · 13%
Q2
Web science4 marks · 13%
Q2
Databases2 marks · 7%
Q1

Mentor insights

Traps to watch for

01
Confusing a network socket with a physical hardware port · Q1
A network socket is an operating system software communication endpoint combining an IP address and a logical port number, not a physical hardware interface like an RJ45 or USB jack.
02
Mentioning software purchase cost in Q2(a) · Q2
The question explicitly demands advantages and disadvantages 'other than cost'; discussing software license prices, development wages, or financial budgets scores zero marks.
03
Assuming VPN encryption protects emergency voice streaming by default · Q3
VPN encapsulation adds packet overhead, introduces jitter, and frequently filters UDP ports used by real-time voice (VoIP), which can degrade emergency audio feeds.

Solutions and tips are GeorgeTheTutor's own; the questions are not reproduced here. Not affiliated with or endorsed by the International Baccalaureate Organization.

Want someone to go through your paper with you?

Our maths mentors took these exams recently. One-to-one, they'll find where your marks are going and fix it before May.

Find your maths mentor More past papers

Worked solutions

IB Computer Science HL · Paper 3

Computer Science, Higher Level

Session: May 2019
Marks: 30
Duration: 1 hour
Calculator: Not allowed
Questions: 4

Q1 Q2 Q3 Q4

Keep your copy of the paper open alongside. We don't reproduce the questions: each part below is labelled to match the May 2019 Global paper, which your school or the IB can provide.

Practice mode

Hide our working until you've had a go.

Questions

Questions 1–4 · 30 marks

Question 1

4 marks

Data tier functions in multi-tier architectures and TCP/IP socket fundamentals

Databases
Networks

(a)

2 marks

Identify two core functions executed by the Data Tier in the EMIS multi-tier architecture:
1) Persistent data storage and transactional retrieval: The data tier executes Data Manipulation Language (DML/SQL) commands to store, query, update, and index emergency incident records and geospatial datasets across relational and spatial database management systems.
2) Enforcing ACID properties and data abstraction: It isolates backend database schemas from application business logic, managing connection pooling, transactional integrity, concurrent locking, and data consistency during simultaneous incident reports.

Mentor tip

Focus on backend data operations: CRUD queries, SQL execution, connection abstraction, and maintaining ACID transaction consistency.

(b)

2 marks

Identify two essential characteristics of a TCP/IP network socket:
1) Endpoint binding to IP address and port: A socket represents an operating system software abstraction uniquely defined by the concatenation of a network IP address and a specific transport-layer port number (e.g., 192.168.1.10:8080).
2) Bidirectional connection-oriented communication: It establishes a full-duplex communication channel between two networked nodes, operating across OSI layers 4 (Transport) and 5 (Session) to stream ordered, acknowledged data packets.

Mentor tip

Highlight the endpoint pairing of an IP address with a port number and its role as a bidirectional transport conduit.

Question 2

8 marks

Software procurement tradeoffs and stateful HTTP connections via URL rewriting

System fundamentals
Web science

(a)(i)

2 marks

Explain one non-cost advantage of purchasing an off-the-shelf emergency response system over a bespoke build:
1) Rapid deployment timeframe: Commercial off-the-shelf software is pre-compiled, tested, and immediately available for installation, enabling Bangbai to deploy emergency capabilities in weeks rather than enduring lengthy multi-year software development cycles where lives could be lost.
2) Proven software reliability: Off-the-shelf packages have undergone rigorous field testing across numerous municipal deployments, offering comprehensive documentation, established user communities, and fewer unforeseen software defects.

Mentor tip

Explain the speed of implementation or extensive real-world testing without mentioning financial costs.

(a)(ii)

2 marks

Explain one non-cost disadvantage of selecting an off-the-shelf emergency response system:
1) Incompatible interfaces with legacy municipal systems: Off-the-shelf platforms are built to generic industry standards and may not integrate with Bangbai's existing bespoke dispatch hardware or specialized emergency sensors without complex integration adapters.
2) Inability to modify closed source code: If Bangbai requires emergency features unique to its local geography (e.g., localized monsoon flood models), municipal engineers cannot alter the underlying codebase without relying on external third-party vendor release roadmaps.

Mentor tip

Focus on rigid closed source code or integration difficulties with existing municipal infrastructure.

(b)

4 marks

Explain two advantages of using URL rewriting to maintain state across stateless HTTP connections:
1) Robust operation independent of client cookie settings: Standard HTTP session persistence relies on browser cookies; however, users frequently disable cookies for privacy reasons or utilize legacy mobile browsers that lack cookie support. URL rewriting embeds the session token directly into the query string parameter, ensuring session continuity across all devices without requiring client-side storage.
2) Enhanced client privacy and forensic auditing: URL rewriting avoids persisting state tokens on the client device filesystem where they could be inspected or harvested by malware. Furthermore, having session tokens explicitly present within server access logs allows network engineers to reconstruct user transaction sequences and trace dropped emergency connections during incident post-mortems.

Mentor tip

Present two distinct advantages: functioning when cookies are disabled or unsupported, and avoiding client-side cookie storage vulnerabilities while enhancing server-side audit logs.

Question 3

6 marks

Technical impediments of VPN tunneling on citizen incident reporting apps

Networks

6 marks

Explain why routing citizen reporting traffic through a VPN reduces the app's operational efficacy:
1) Masking true origin IP addresses: A VPN routes packets through an intermediary external relay server, replacing the citizen's authentic IP address with the exit node IP address. If the smartphone's internal GPS is disabled or indoors, the emergency dispatch server cannot perform network-based geolocation or verify the physical location of the reporter.
2) Latency overhead and connection drops: Encapsulating, tunneling, and encrypting network packets increases protocol overhead and packet header sizes, introducing transmission latency and elevating the probability of dropped connections over volatile cellular links during crises.
3) Port blocking and protocol filtering: Commercial VPN providers often apply restrictive filtering rules or block UDP and non-standard communication ports, inadvertently disabling VoIP streams, video feeds, and real-time push telemetry required for emergency assessment.
4) Obscuring geographic telemetry validation: Automated dispatch engines cross-reference reported GPS coordinates against the caller's IP geolocation to detect fraudulent calls; a mismatch caused by a foreign VPN exit node can delay dispatch triage.
5) Anonymity-induced abuse: Perceived absolute anonymity can encourage malicious actors to submit false reports, consuming emergency dispatch personnel and diverting finite first responders away from authentic life-threatening incidents.
6) Third-party logging liabilities: Unvetted commercial VPN nodes can log sensitive user data and metadata, compromising the confidentiality of whistleblowers reporting dangerous criminal incidents.

Mentor tip

Explain how VPN masking breaks IP-based geolocation, adds encryption overhead/latency, blocks essential streaming ports, and complicates hoax verification.

Question 4

12 marks

Comparative evaluation of server load balancing algorithms for Bangbai EMIS

Networks

12 marks

Evaluate the appropriateness of Client-Side Random versus Source IP Hash load balancing for the Bangbai EMIS citizen reporting server cluster:
1) Operational mechanisms:
- Client-Side Random: The client application retrieves an inventory of available server IP addresses and randomly selects an endpoint for each request without consulting a centralized server-side distributor.
- Source IP Hash: A load balancer computes a mathematical hash from the client's source IP address and destination address, modulo the number of active cluster nodes ( $ServerIndex = Hash (IP_{source}) (mod N)$ ), routing all packets from that IP to a deterministic server.
2) Session persistence and stateful operations:
- Source IP Hash maintains intrinsic session stickiness: subsequent requests from the same user land consistently on the same backend node, preserving active in-memory incident reporting drafts and file uploads without requiring clustered session synchronization.
- Client-Side Random lacks session affinity: consecutive requests may land on different servers, terminating half-completed reporting forms unless expensive shared distributed caching is implemented.
3) Vulnerability to asymmetric traffic and NAT aggregation:
- Source IP Hash suffers from severe load imbalance when hundreds of citizens report from behind a single shared cellular carrier NAT or municipal proxy gateway; all requests share an identical public IP, routing all traffic to a single server and creating bottleneck hotspots.
- Client-Side Random distributes requests evenly across all cluster nodes regardless of client IP concentration, naturally preventing single-server starvation caused by carrier NAT.
4) Security, denial-of-service, and single points of failure:
- Client-Side Random exposes all internal cluster server IP addresses directly to the public client, allowing malicious actors to target specific backend nodes with directed Denial of Service (DoS) floods.
- Source IP Hash operates behind a shielded reverse proxy that conceals backend topologies. However, that centralized reverse proxy represents a single point of failure (SPOF) unless deployed with redundant hardware pairs.
5) Cost, implementation complexity, and synthesis:
- Client-Side Random requires zero expensive hardware load balancers, making it cheap to implement, but its lack of persistence and heightened security vulnerability make it inadequate for high-reliability emergency systems.
- Recommendation: Source IP Hash deployed on an active-passive hardware load balancer pair is far superior for Bangbai EMIS, providing critical session continuity for emergency incident uploads, provided that consistent hashing algorithms are employed to gracefully accommodate cluster node scaling.

Node = Hash (IP_{source}) (mod N)

Mentor tip

Compare both algorithms across session persistence, traffic distribution under carrier NAT, security/DoS attack exposure, single points of failure, and conclude with a justified recommendation.

Our analysis

Where the marks are

Each part sorted by the topic it mostly tests. Networks carries the most marks on this paper.

Networks20 marks · 67%
Q1, Q3, Q4
System fundamentals4 marks · 13%
Q2
Web science4 marks · 13%
Q2
Databases2 marks · 7%
Q1

Mentor insights

Traps to watch for

01
Confusing a network socket with a physical hardware port · Q1
A network socket is an operating system software communication endpoint combining an IP address and a logical port number, not a physical hardware interface like an RJ45 or USB jack.
02
Mentioning software purchase cost in Q2(a) · Q2
The question explicitly demands advantages and disadvantages 'other than cost'; discussing software license prices, development wages, or financial budgets scores zero marks.
03
Assuming VPN encryption protects emergency voice streaming by default · Q3
VPN encapsulation adds packet overhead, introduces jitter, and frequently filters UDP ports used by real-time voice (VoIP), which can degrade emergency audio feeds.

Solutions and tips are GeorgeTheTutor's own; the questions are not reproduced here. Not affiliated with or endorsed by the International Baccalaureate Organization.

Want someone to go through your paper with you?

Our maths mentors took these exams recently. One-to-one, they'll find where your marks are going and fix it before May.

Find your maths mentor More past papers