IB Computer Science A3.4: Alternative databases and data warehouses (HL only)

New IB Computer Science syllabus · first exams May 2027 · Theme A · HL only

A3.4 goes beyond relational databases. It covers other database models, how organisations analyse huge amounts of historical data, and how databases are spread across many machines.

Key points

A3.4.1 Database models

• NoSQL ("not only SQL"): non-relational, with a flexible schema. Includes document, key-value, column-family and graph stores. Used for social media posts.
• Cloud: a database run as a managed service. Used for SaaS applications.
• Spatial: built for geographic data and queries. Used for GIS and ride-sharing routes.
• In-memory: data held in RAM for very low latency. Used for real-time leaderboards and analytics.

A3.4.2 Data warehouses

A data warehouse is a large store that brings together cleaned data from many sources for analysis. Its data is:

• subject-oriented: organised around business subjects
• integrated: consistent across sources
• time-variant: every record is dated
• non-volatile and append-only: history is never overwritten
• optimised for query performance: usually denormalized

A3.4.3 OLAP and data mining

• OLAP (online analytical processing): explores data from several angles at once. Its operations are slice, dice, drill-down, roll-up and pivot. Don't confuse it with OLTP, which handles everyday transactions.
• Data mining techniques you must know:
  • classification
  • clustering
  • regression
  • association rule discovery
  • sequential pattern discovery
  • anomaly detection

These link to A4 Machine learning.

A3.4.4 Distributed databases

A distributed database is spread across several locations but appears to users as one database. Its features are:

• concurrency control
• data consistency
• data partitioning
• security
• distribution and location transparency
• fault tolerance
• global query processing
• replication
• scalability

ACID still matters for reliable transactions, and keeping the copies consistent is the central challenge.

Exam-style question

Explain three of the six characteristics of data warehouse data and why each supports business intelligence. [6]

Mark scheme

Award [2] per characteristic ([1] for describing it, [1] for the business intelligence benefit), any three of:

• Time-variant: every record carries a date, which enables trend analysis and year-on-year comparisons.
• Non-volatile: loaded data is never changed, so historical reports can be reproduced.
• Integrated: data from many sources is put into one consistent format, giving a "single source of truth".
• Subject-oriented: organised around business subjects, which supports decisions across business areas.
• Append-only: new data is added without overwriting old data, preserving history.
• Optimised for queries: denormalized design, pre-calculated totals and indexes allow fast analysis of huge datasets.

Common mistakes that cost marks

• Saying "NoSQL means no SQL". It means not only SQL. The defining feature is the non-relational structure, not the query language.
• Mixing up OLAP and OLTP. OLAP is for analysis; OLTP is for everyday transactions.
• Naming data mining techniques without saying what each finds in a large dataset.

Frequently asked questions

Is A3.4 on the SL syllabus?
No. The whole of A3.4 is HL only.

Which database models do I need to know?
The four named in the guide: NoSQL, cloud, spatial and in-memory.

Go further

• Practise A3.4 with instant marking on ExamHaus. → examhaus.com
• Full A3.4 presentation and question bank with mark scheme on CS Exam Ready. → csexamready.com/resources/ib-diploma
• IB CS mock exams: Paper 1 & 2, SL & HL. → examhaus.com/mock-exams