SD Core

SQL vs NoSQL

Pick the store based on access patterns — ACID transactions vs horizontal write scale and flexible schema.

Interview tip Lead with a 30-second definition, then one real system example and name 2–3 designs where SQL vs NoSQL is non-negotiable.

① What it is (30 seconds)

Pick the store based on access patterns — ACID transactions vs horizontal write scale and flexible schema.

② How it works in system design

SQL (PostgreSQL): joins, constraints, transactions. NoSQL families: wide-column (Cassandra writes), document (Mongo flexible JSON), key-value (Redis speed), search (Elasticsearch text).
Typical placement
ClientEdge / GatewaySQLServicesData stores

③ Concrete system design example

Scenario: E-commerce: PostgreSQL for orders + payments (ACID). Redis for cart session. Elasticsearch for product search. S3 for product images.

④ Important interview Q&A

QuestionAnswer
When SQL wins?Financial ledger, inventory, relationships needing joins and strong constraints.
When Cassandra wins?High write throughput, time-series, tolerate eventual consistency per row.
Can you mix?Polyglot persistence — each service owns its store optimized for its pattern.

⑤ Seen in these system designs

In interviews, after explaining the concept, say: "This shows up directly in …" and link two designs.

⑥ Revision checklist

  • Access pattern first
  • ACID needs stated
  • Polyglot example
  • No buzzword-only picks
sqlnosqldatabase