Learn SQL in Action

Kaufman skill map untuk menguasai SQL secara efektif, dengan target skill produksi, operating model SQL, feedback loop, deliberate practice, dan mental model engineer-level.

Fondasi relational model untuk SQL production: relation, tuple, attribute, domain, predicate, key, set vs bag, NULL, three-valued logic, dan relational reasoning.

Mental model database engine untuk memahami perjalanan SQL dari text menjadi result: parser, binder, rewriter, optimizer, executor, storage engine, buffer pool, WAL, lock manager, MVCC, dan catalog.

SQL syntax sebagai intent deklaratif: SELECT, FROM, WHERE, GROUP BY, HAVING, ORDER BY, LIMIT/FETCH, logical processing order, alias scope, expression evaluation, dan bug semantic umum.

Schema design as invariant engineering: tables, keys, constraints, defaults, identity, generated columns, referential integrity, naming, and production review discipline.

Data types as domain contracts: numeric precision, temporal correctness, text/collation, boolean modelling, UUIDs, enums, domains, binary data, coercion, and production failure modes.

Semantik INSERT, UPDATE, DELETE, dan MERGE untuk sistem produksi: mutasi data, upsert, returning/output, write amplification, idempotency, optimistic update, lost update prevention, dan desain DML yang aman.

Filtering, predicates, three-valued logic, NULL traps, IN/EXISTS, BETWEEN, LIKE, regex, sargability, implicit cast, dan desain predicate yang benar untuk sistem produksi.

Join sebagai mekanisme menggabungkan fakta dan relationship secara benar, termasuk inner join, outer join, semi join, anti join, fan-out, null-extension, dan debugging cardinality.

Aggregation, grouping, dan measure correctness untuk SQL produksi, termasuk COUNT variants, SUM/AVG, GROUP BY, HAVING, distinct aggregation, conditional aggregation, rollup, cube, grouping sets, dan metric traps.

Subqueries, CTEs, and query composition for production SQL: scalar/table subqueries, correlated queries, derived tables, recursive CTEs, LATERAL/APPLY, materialization behavior, and composable query contracts.

Production-grade window functions: partitioning, ordering, frames, ranking, running totals, latest-row selection, gaps-and-islands, sessionization, deduplication, audit timelines, and performance reasoning.

Production-grade set operations and relational differencing: UNION, INTERSECT, EXCEPT, ALL vs DISTINCT, reconciliation, delta detection, data comparison, audit validation, and correctness traps.

First-principles index literacy: B-tree, hash, bitmap, BRIN/zonemap, covering indexes, composite order, selectivity, clustering, write cost, and production index trade-offs.

Practical index design through sargability, access paths, predicate shape, composite index order, covering indexes, partial indexes, expression indexes, and production workload reasoning.

Query plan literacy for production SQL: EXPLAIN, actual plans, scans, joins, sorts, aggregates, cost estimates, cardinality, plan debugging, and optimizer reasoning.

Cardinality, statistics, and cost model failures for production SQL: row estimates, histograms, skew, correlation, parameter-sensitive plans, stale stats, and practical remediation.

Transactions, ACID, and unit-of-work design for production SQL: transaction boundaries, commit/rollback, savepoints, idempotency, retries, invariants, and application/database consistency.

Isolation levels, MVCC, locks, and concurrency anomalies for production SQL: visibility rules, lost updates, write skew, phantoms, serializability, and anomaly testing.