Zooming out from a single database to enterprise-scale data concepts: threats to data quality, data warehousing, data mining, big data, and NoSQL.
Threats to data quality:
| Threat | Description |
|---|---|
| Corrupted data | Data that has been damaged/altered unintentionally, often becoming unusable or inaccurate |
| Outdated data | Data that was once accurate but is no longer current |
| Invalidated data | Data that hasn't passed proper validation checks and may not meet expected rules |
| Vulnerability | Weaknesses that expose data to attack, e.g. SQL injection (malicious code inserted via input) or malware infections |
Purpose and uses: data warehousing consolidates historical data for large-scale analysis; data mining extracts actionable insights and patterns from that consolidated data.
Big data โ sources include social media, activity-generated data (e.g. app usage logs), server logs and audit trails, and location-based data (GPS/LBS).
| Comparison | SQL (relational) | NoSQL |
|---|---|---|
| Structure | Fixed schema, structured tables with defined relationships | Flexible, often schema-less โ good for varied/changing data shapes |
| Scalability | Traditionally scales vertically (more powerful single server) | Often designed to scale horizontally (many distributed servers) โ well suited to big data |
| Best for | Structured data with clear relationships, requiring complex queries/joins | Large volumes of unstructured/semi-structured, rapidly changing data |
๐ก Exam Tip
The SQL vs NoSQL comparison is a common essay-style question โ anchor your answer around structure (fixed schema vs flexible) and scalability (vertical vs horizontal), which are the two clearest, most defensible distinguishing factors.