malisper/pgrust

Postgres rewritten in Rust, now faster than Postgres and Clickhouse

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 1 hour ago
Type:Server / PlatformCategory(s):Relational & SQL DatabasesDatabases & Storage
Added to GitGenius on September 13th, 2026
Created on April 20th, 2026
Open Issues & Pull Requests: 6 (+0)
GitHub issues: Enabled
Number of forks: 191
Total Stargazers: 5,099 (+1)
Total Subscribers: 45 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 2.1 days
Mean response time: 10.1 days
90th percentile: 33.9 days
Tracked items: 72

How this project is maintained

Practically every issue opened in the past year has drawn a reply. 92% of issues opened in the past year have since been closed. Three people close 99% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 6
New in 7 days: 1
Closed in 7 days: 0
Avg open age: 31 days
Stale 30+ days: 2
Stale 90+ days: 0

Recent activity

Opened in 7 days: 1
Closed in 7 days: 0
Comments in 7 days: 1
Events in 7 days: 1

Top labels

  • enhancement (2)

Most active issues this week

Sign in to see which issues are moving.

Detailed Description

pgrust is a Postgres rewrite in Rust that aims to show what Postgres would look like if built with modern systems programming practices and AI-assisted development.

The project addresses fundamental reliability and performance issues in Postgres by reimplementing the database engine from scratch in Rust. The tool is wire-compatible and SQL-dialect compatible with Postgres, meaning existing applications can connect to it without modification. It passes the complete Postgres regression test suite. The architecture introduces several improvements over the original: a vectorized, JIT-compiled query executor; a thread-based concurrency model; a query scheduler that prevents individual queries from destabilizing the database; and a built-in out-of-memory killer that gives the database control over memory pressure rather than relying on the operating system. These changes directly target reliability problems identified as common causes of Postgres outages.

Adoption requires understanding that pgrust is explicitly not production-ready and should not be used with data that matters. The project is in active development with testing and reliability as the stated priority. It performs significantly faster than Postgres and ClickHouse on analytical workloads when using its built-in columnar storage format. However, existing PostgreSQL extensions do not work, as there is no stable extension ABI yet, though some bundled contrib modules have been ported. The JIT compiler is specifically tuned for Graviton4 processors; the tool runs on other platforms but without equivalent performance gains. This makes pgrust suitable for developers interested in exploring next-generation database architecture or evaluating performance on compatible hardware, but not for production deployments requiring stability or extension support.

Development activity shows active implementation of core database functionality with comprehensive test coverage against the Postgres regression suite. The project maintains a public roadmap and communication channels for community engagement.