PostgreSQL completed a SELECT query against one million records in under 1 millisecond on Linux in 2024. MySQL needed between 9 and 12 milliseconds for the same operation. This article compiles benchmark results from peer-reviewed research, engineering blogs, and developer surveys published between 2024 and early 2026 to show where each database actually performs on Linux production hardware.
MySQL vs PostgreSQL Statistics: Key Numbers for 2026
- PostgreSQL is used by 51.9% of professional developers, versus 39.4% for MySQL, according to the Stack Overflow Developer Survey 2024.
- PostgreSQL completed SELECT queries on 1 million records in 0.6–0.8 ms, roughly 13 times faster than MySQL’s 9–12 ms, in a 2024 MDPI study.
- PostgreSQL sustained 21,338 single-row INSERT operations per second on Ubuntu 24.04; MySQL capped at 4,383 QPS in the same January 2026 test.
- Across 17 Sysbench tests on Ubuntu Linux, PostgreSQL posted 2.3 times lower median latency overall and 3.5 times lower latency on write operations.
- MySQL recorded up to 21% higher peak TPS on simple single-table OLTP workloads, the one benchmark category where it consistently leads PostgreSQL.
Which Database Do Linux Developers Actually Use?
PostgreSQL crossed 45% adoption in 2023 and reached 48.7% across all developers in 2024. Among professional developers specifically, the gap widens: PostgreSQL at 51.9% versus MySQL at 39.4%.
The picture was different in 2018, when MySQL led at 59% and PostgreSQL sat at 33%. MySQL retains an edge among developers learning to code, where it still leads at 45% against PostgreSQL’s 33%, partly due to its presence in LAMP stack setups and shared hosting environments.
Source: Stack Overflow Developer Survey 2024 (65,000+ respondents, 185 countries)
| Database | All Developers | Professional Developers | Learning to Code |
|---|---|---|---|
| PostgreSQL | 48.7% | 51.9% | 33% |
| MySQL | 40.3% | 39.4% | 45% |
| SQLite | 33.1% | 32.1% | 36% |
Source: Stack Overflow Developer Survey 2024
PostgreSQL vs MySQL SELECT Performance on Linux at 1 Million Records
A 2024 benchmarking study published in MDPI’s Future Internet tested SELECT operations against both databases on workloads modeled on continuous authentication systems—high-frequency reads against large behavioral datasets. PostgreSQL’s MVCC implementation and query planner were identified as the primary factors behind the gap.
Source: Salunke & Ouda, Future Internet, MDPI, October 2024
| Operation | PostgreSQL | MySQL | Faster |
|---|---|---|---|
| Primary SELECT (1M records) | 0.6–0.8 ms | 9–12 ms | PostgreSQL (~13x) |
| Complex SELECT (multi-condition) | Lower latency | Higher latency | PostgreSQL |
Source: Salunke & Ouda, Future Internet, MDPI, 2024
Sysbench Benchmark Results: PostgreSQL 15.5 vs MySQL 8.0.35 on Ubuntu Linux
DoltHub runs nightly Sysbench benchmarks on AWS EC2 (m5a.xlarge: 4 AMD EPYC 7000 cores, 16GB RAM, Ubuntu Linux). Their July 2024 comparison of PostgreSQL 15.5 against MySQL 8.0.35—both at default configuration—covered 17 test types. MySQL led on one: index_join at 1.37 ms versus PostgreSQL’s 1.96 ms. The two databases were essentially tied on select_random_ranges.
Across the remaining 15 tests, PostgreSQL posted lower latency. Results are stable within 2% variance across nightly runs, which makes these figures a reliable reference point.
Source: DoltHub Blog, July 2024
| Sysbench Test | PostgreSQL 15.5 (ms) | MySQL 8.0.35 (ms) | Faster |
|---|---|---|---|
| oltp_point_select | 0.14 | 0.18 | PostgreSQL |
| oltp_read_only | 2.52 | 3.49 | PostgreSQL |
| oltp_read_write | 4.33 | 8.58 | PostgreSQL |
| oltp_insert | 1.03 | 3.82 | PostgreSQL |
| oltp_update_index | 1.06 | 3.89 | PostgreSQL |
| oltp_write_only | 1.67 | 5.47 | PostgreSQL |
| groupby_scan | 5.28 | 14.46 | PostgreSQL |
| index_scan | 18.28 | 34.33 | PostgreSQL |
| table_scan | 18.61 | 34.95 | PostgreSQL |
| index_join | 1.96 | 1.37 | MySQL |
| select_random_ranges | 0.41 | 0.39 | MySQL (~equal) |
Source: DoltHub Blog, July 2024
MySQL vs PostgreSQL Write Throughput on Ubuntu 24.04: 2026 Data
A January 2026 benchmark by BinaryIgor tested MySQL 9.5 and PostgreSQL 18.1 on identical hardware: AMD Ryzen 7 PRO 7840U, 8 cores, 32GB RAM, Ubuntu 24.04.3 LTS, Samsung NVMe SSD. Each database ran in Docker with 16GB RAM and 8 CPU allocation. MySQL received 128 connections; PostgreSQL received 64—MySQL empirically benefits from higher connection counts on write-heavy loads, so the test was set up to give it that advantage.
PostgreSQL still handled 21,338 single-row inserts per second. MySQL capped at 4,383. At the 99th percentile, PostgreSQL’s insert latency was 4.0 ms against MySQL’s 42.7 ms.
Source: BinaryIgor, January 2026
| Operation | MySQL 9.5 QPS | MySQL Mean (ms) | PostgreSQL 18.1 QPS | PostgreSQL Mean (ms) |
|---|---|---|---|---|
| Single-row INSERT | 4,383 | 26.8 | 21,338 | 2.39 |
| Batch INSERT 100 rows (items) | 200 | 26.5 | 211 | 4.1 |
| Batch INSERT 100 rows (orders) | 1,883 | 51.6 | 3,535 | 14.7 |
Source: BinaryIgor, January 2026
When Does MySQL Outperform PostgreSQL on Linux?
A 2025 study in the International Journal of Modern Computer Science and IT Innovations ran both databases through SysBench, TPC-C, and TPC-H on a dedicated physical server. MySQL posted up to 21% higher peak TPS on simple OLTP—single-table, high-concurrency reads and writes at moderate connection counts. This pattern aligns with MySQL’s buffer pool and index traversal behavior, which handles certain single-row join patterns efficiently.
Once workloads involve multiple joined tables, aggregations, or analytical queries, PostgreSQL leads across all three test categories. A December 2024 test at Vonng found PostgreSQL 16 matched or exceeded MySQL 9.0 on sysbench/wrk workloads simply by enabling prepared statements—no other tuning needed.
| Benchmark | Workload Type | Faster Database | Margin |
|---|---|---|---|
| SysBench OLTP | Single-table, high-concurrency reads/writes | MySQL | Up to 21% higher peak TPS |
| TPC-C | Multi-table complex transactions | PostgreSQL | PostgreSQL outperforms |
| TPC-H (22 queries) | Decision-support / analytical | PostgreSQL | PostgreSQL outperforms |
Source: International Journal of Modern Computer Science and IT Innovations, 2025
FAQs
Is PostgreSQL faster than MySQL on Linux?
On most workloads, yes. PostgreSQL leads on reads, writes, complex queries, and analytical operations. MySQL posts up to 21% higher TPS only on simple single-table OLTP under moderate concurrency.
Which database do professional developers prefer in 2024?
PostgreSQL. The Stack Overflow Developer Survey 2024 recorded PostgreSQL at 51.9% among professional developers versus MySQL at 39.4%, reversing the gap that existed in 2018.
How much faster is PostgreSQL for INSERT operations?
In January 2026 testing on Ubuntu 24.04, PostgreSQL handled 21,338 single-row inserts per second versus MySQL’s 4,383—a 4.87x throughput difference under identical hardware conditions.
What benchmark did MySQL win in the DoltHub Sysbench tests?
MySQL was faster on index_join at 1.37 ms versus PostgreSQL’s 1.96 ms, and essentially tied on select_random_ranges. PostgreSQL led on the other 15 of 17 test types.
What hardware was used in the 2026 MySQL vs PostgreSQL benchmark?
AMD Ryzen 7 PRO 7840U (8 cores), 32GB RAM, Samsung NVMe SSD, Ubuntu 24.04.3 LTS. Both databases ran in Docker limited to 16GB RAM and 8 CPUs. MySQL received 128 connections; PostgreSQL received 64.
