Database Indexing Benchmark

Comparing query performance with and without indexes

POSTGRES Improvement

-3.1%

Average p95 latency reduction with indexes

MYSQL Improvement

31.0%

Average p95 latency reduction with indexes

Scenarios Tested

9

Different query patterns benchmarked

Databases Compared

2

POSTGRES, MYSQL

P95 Response Time Comparison (ms)

Internal Database Query Time (p95 ms)

Throughput Comparison (requests/sec)

Latency Percentiles by Database

POSTGRES - All Percentiles (With Index)

MYSQL - All Percentiles (With Index)

Per-Scenario Improvement (HTTP)

Scenario	SQL Query	POSTGRES No Index (ms)	POSTGRES With Index (ms)	Improvement	MYSQL No Index (ms)	MYSQL With Index (ms)	Improvement
aggregation	SELECT COUNT(*) as count FROM products WHERE category = ?	15.38	12.62	+17.9%	17.44	15.75	+9.7%
composite-filter	SELECT * FROM users WHERE country = ? AND status = ? LIMIT 100	12.27	17.78	-44.9%	22.31	19.78	+11.3%
join-query	SELECT t.*, u.name as user_name, p.name as product_name FROM transactions t JOIN users u ON t.user_id = u.id JOIN products p ON t.product_id = p.id WHERE t.user_id = ? ORDER BY t.created_at DESC LIMIT 50	44.98	44.67	+0.7%	16.33	15.39	+5.8%
pagination	SELECT * FROM transactions ORDER BY id LIMIT ? OFFSET ?	15.60	17.16	-10.0%	17.52	16.83	+3.9%
point-lookup	SELECT * FROM users WHERE email = ?	10.42	11.79	-13.2%	14.61	10.93	+25.1%
prefix-search	SELECT * FROM users WHERE name LIKE ? LIMIT 50	19.67	17.27	+12.2%	24.27	16.95	+30.1%
range-query	SELECT * FROM users WHERE created_at BETWEEN ? AND ? LIMIT 100	14.06	18.17	-29.2%	59.22	21.96	+62.9%
sorting	SELECT * FROM products ORDER BY price DESC LIMIT 20	13.11	14.04	-7.1%	19.54	13.34	+31.8%
status-filter	SELECT * FROM transactions WHERE status = ? AND created_at BETWEEN ? AND ? ORDER BY created_at DESC LIMIT 100	34.77	18.84	+45.8%	1560.99	22.82	+98.5%

Index Impact by Scenario