DBMS & Warehousing GATE DA: Syllabus, Books, PYQs

Most GATE DA aspirants prepare DBMS just fine — Korth is Korth, the ER-to-schema drills are the same as GATE CS, and SQL predict-the-output is a well-understood genre. The problem is the other half: data warehousing. There is no single textbook that maps cleanly to the warehousing syllabus, so aspirants either skip it or read scattered web articles. Both options leave marks on the table. This page covers both halves — DBMS and warehousing — with the books, the curated reading strategy for warehousing, and a study sequence that treats them as the roughly equal halves they are.

The Two-Half Split

Official Syllabus

DBMS

• ER model
• Relational model — relational algebra, tuple calculus, SQL
• Integrity constraints, normal form
• File organization, indexing
• Data types

Data Warehousing

• Data transformation — normalization, discretization, sampling, compression
• Warehouse modelling — schema for multidimensional data models, concept hierarchies, measures: categorization and computations

The DBMS list is essentially GATE CS DBMS — Korth maps cleanly. The warehousing list sits across data-mining and warehousing literature, which is why no single textbook covers it end-to-end.

Books and Resources

Full book list: GATE DA books and resources guide.

DBMS — Topic by Topic

ER Model

Entities, attributes (simple, composite, multi-valued, derived), relationships, cardinality, participation constraints, weak entities. Practise converting ER diagrams to relational schemas — both ways.

Relational Algebra and Tuple Calculus

Selection, projection, set operations, joins (theta, equi-, natural, outer), division. Tuple calculus — predicate formulation. Fluency here makes SQL questions much faster.

SQL

The single highest-yield DBMS topic. Cover: SELECT with WHERE, GROUP BY, HAVING, ORDER BY; all join types; aggregate functions (COUNT, SUM, AVG, MIN, MAX); nested queries (correlated and non-correlated); set operations (UNION, INTERSECT, EXCEPT); CASE expressions. PYQ pattern: a small sample table with an SQL query — predict the output rows. Treat it like a Python predict-output question.

Integrity Constraints

Primary key, foreign key, unique, not null, check. Understand referential integrity and cascading deletes / updates.

Normal Forms

1NF, 2NF, 3NF, BCNF. Functional dependencies, candidate key identification, BCNF decomposition, dependency preservation trade-off. Decomposition questions are more common than definition questions.

File Organization and Indexing

Heap, sequential and hashed file organization. B+ tree indexing — structure, search, insertion, deletion at a conceptual level; dense vs sparse indexes; primary vs secondary indexes.

Data Warehousing — Topic by Topic

Data Transformation

Normalization (scaling to 0–1, z-score), discretization (binning — equal-width vs equal-frequency), sampling (random, stratified), compression at a conceptual level. These are pre-processing techniques that bridge into the machine learning syllabus.

Schemas — Star, Snowflake, Fact Constellation

Star schema: one fact table linked to denormalised dimension tables. Snowflake: dimensions further normalised into sub-dimensions. Fact constellation: multiple fact tables sharing dimensions. Be able to identify a schema from a diagram and articulate trade-offs (query simplicity vs storage/update efficiency).

Concept Hierarchies and Measures

Concept hierarchies organise dimension attributes into levels (day → month → quarter → year). Measures are aggregated values in the fact table (sum of sales, count of orders) — additive, semi-additive (averages, ratios), or non-additive (percentages). OLAP operations — roll-up, drill-down, slice, dice, pivot — sit on top of these.

Past-Paper Patterns

DBMS and warehousing PYQs since 2024 cluster around five types:

1. SQL query output prediction on a small sample table
2. Functional dependencies and normal-form decomposition
3. ER-to-relational conversion
4. Schema identification (star vs snowflake)
5. Data-transformation conceptual questions (normalization, discretization)

Solve every DBMS and warehousing PYQ from official GATE DA 2024 and 2025 papers.

What to Skip

• Transaction management and concurrency control beyond basics — heavy in GATE CS, lighter in DA.
• Distributed databases — distributed concurrency, distributed queries, CAP theorem details.
• Recovery and logging beyond conceptual exposure.
• Data-mining algorithms (Apriori, FP-growth) — mining ≠ warehousing; k-means belongs to ML.
• Big data platforms (Hadoop, Spark) and NoSQL internals.

Study Sequence

Weeks 1–3: DBMS

1. ER model, relational algebra, tuple calculus — Korth chapters 1–4.
2. SQL — Korth chapters 5–6; output-prediction practice on sample tables.
3. Integrity constraints.
4. Normal forms — functional dependencies, BCNF decomposition.
5. File organization and B+ tree indexing.

Weeks 4–5: Data Warehousing

1. Data transformation — normalization, discretization, sampling, compression (Han / Kamber / Pei data-preprocessing chapters).
2. Warehouse modelling — star, snowflake, fact constellation; identify from diagrams.
3. Concept hierarchies, measures, OLAP operations.

Weeks 6–7: PYQs and Revision

1. Every DBMS and warehousing PYQ from GATE DA 2024 and 2025.
2. 2–3 topic-wise tests from The ML Hub GATE DA test series.
3. Finalise the SQL pattern sheet and schema-comparison reference.

Traps to Avoid

• Skipping warehousing because "there is no textbook". Roughly half the section is warehousing. Combine Han / Kamber / Pei with mentor notes.
• Confusing star and snowflake schemas. Star = denormalised dimensions; snowflake = normalised. Know the trade-offs.
• Memorising normal-form definitions without practising decomposition. Decomposition is the more common PYQ pattern.
• Over-studying transactions. Heavy in GATE CS, lighter in DA.
• Confusing data-mining and warehousing. Apriori is out of scope; OLAP and schemas are in.

DBMS & Warehousing in The ML Hub's Course

The DBMS block in The ML Hub's GATE DA course covers the full DBMS syllabus via Korth, plus curated warehousing material that bridges the gap no single textbook fills. Lectures include SQL execution traces, normal-form decomposition drills, and schema identification exercises. The test series has topic-wise packs for SQL, schemas, and normal forms. The study schedule places DBMS after Python & DSA, sharing the predict-output discipline.

FAQs

Can I skip warehousing in GATE DA?

No. Warehousing is roughly half the section. Use Han / Kamber / Pei's data-mining textbook (warehouse-modelling chapters) plus curated mentor notes.

Which book for GATE DA DBMS?

Korth / Silberschatz / Sudarshan — Database System Concepts. Same canonical textbook as GATE CS.

Is OLAP in the GATE DA syllabus?

OLAP operations (roll-up, drill-down, slice, dice, pivot) fall under "warehouse modelling, schemas, concept hierarchies and measures" — so yes, conceptually in scope.

Is "normalization" in GATE DA about normal forms or data scaling?

Both. Normal forms (1NF–BCNF) sit under DBMS. Normalization as data transformation (0–1 scaling, z-score) sits under warehousing. Both are in the syllabus.

Related Guides

Subjects that share question machinery: Python & DSA (SQL traces echo predict-output) and Machine Learning (data transformation feeds into preprocessing). Full subject map: GATE DA books and resources · GATE DA syllabus 2027 guide.