Tutorial 3: Databases and Big Data

COMP200 · Unit 4 · Section 3

Objectives

Relational databases organize structured records and support constraints and transactions. Big-data systems address volume, velocity, variety, and distributed processing. Scale does not excuse poor data governance: quality, lineage, access, retention, and meaning remain essential.

source -> ingest -> validate -> store -> query or analyze -> decision

Exercises

  1. Choose relational or distributed storage for three workloads.
  2. Define data-quality checks.
  3. Trace a query through an index.

Self-check

  1. What does a transaction protect?
  2. Why use an index?
  3. What is data lineage?

Self-Check Quiz

1. What is a relational table?

AnswerA structured collection of records and fields governed by a relational schema.

2. Does big data automatically mean better decisions?

AnswerNo. Decisions depend on quality, relevance, context, governance, and appropriate analysis.

Homework

  1. Design storage for transaction records and event streams.
  2. Specify quality and access controls.
  3. Explain one consistency trade-off.
Sample answerTransactions fit a relational database with constraints and ACID behavior; high-volume events may use distributed ingestion and analytical storage. Both need validation, lineage, authorization, retention, and monitoring. Event systems may trade immediate consistency for scale.