database-query-profiling

Search for queries executed inside loops or list iterations and check for missing eager-loading calls such as select_related in Django, joinedload in SQLAlchemy, includes in ActiveRecord, or include in Prisma. Each missing call causes one extra query per iterated row.

Examine columns used in WHERE, JOIN ON, ORDER BY, and GROUP BY clauses, then compare them against existing indexes. Run EXPLAIN or EXPLAIN ANALYZE to confirm sequential scans, and create composite indexes matching multi-column filter patterns.

Yes, queries can be analyzed statically from source code, including N+1 detection and index recommendations based on query patterns. However, EXPLAIN plan analysis and actual performance measurement require a live database connection.

PostgreSQL, MySQL/MariaDB, and SQLite are supported for EXPLAIN analysis. ORM coverage includes Django, SQLAlchemy, ActiveRecord, Prisma, TypeORM, and Sequelize for tracing generated SQL and eager-loading fixes.

Subqueries in IN clauses can prevent the optimizer from using efficient join strategies and indexes. Rewriting them as INNER JOINs with DISTINCT typically produces better execution plans, which you can verify by comparing EXPLAIN output before and after.