NoSQL and Search for Backend Developers
NoSQL and search systems are data stores that do not use the relational table model. Document stores such as MongoDB hold flexible JSON-like records, key-value stores such as Redis give very fast lookups, and search engines such as Elasticsearch build inverted indexes for full-text queries and relevance ranking.
Stage: Databases · Part of the Backend Developer roadmap · Last reviewed 19 November 2026
What to learn in NoSQL and Search
- Model a document and decide between embedding and referencing
- Use Redis commands for strings, hashes and expiring keys
- Explain what an inverted index is and why it makes text search fast
- Pick a store from the access pattern of a feature
- Describe the consistency trade-offs of distributed NoSQL systems
NoSQL and Search interview questions
When is a document database a better fit than SQL?
Choose a document database when records have varying or evolving shapes, are usually read and written as a whole, and have few cross-record joins, such as product catalogues or content. If you need complex joins and strong relational integrity, SQL is usually better.
Why use Elasticsearch instead of a SQL LIKE query?
LIKE with a leading wildcard scans every row and has no concept of relevance. Elasticsearch tokenises text into an inverted index mapping terms to documents, so it can search large text quickly, rank results, handle typos and facets. It complements the main database rather than replacing it.
Where NoSQL and Search fits in the Backend Developer roadmap
NoSQL and Search belongs to the Databases stage (beginner level), which takes about 5 hours. Model, query and tune data stores.
Other units in this stage
Back to the full Backend Developer roadmap