apache/arrow-rs

Health v2 · maintenance only

Adoption not part of the score

3592 stars · 1281 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

The official Rust implementation of Apache Arrow, providing the in-memory columnar format, and of Apache Parquet for columnar file reading and writing. It ships as a set of crates including core arrow arrays, compute kernels, IPC, CSV/JSON support, and the parquet crate.

Use cases

• read and write parquet files in rust
• columnar in-memory data processing
• zero-copy data interchange between processes
• build analytical query engines in rust
• convert csv and json to arrow record batches
• efficient data transfer between rust and python via arrow

When to choose

• you need fast, memory-efficient columnar data processing in Rust
• you are reading or writing Parquet files
• you want zero-copy interoperability with other Arrow implementations like PyArrow
• you are building data infrastructure such as query engines or ETL pipelines in Rust

When to avoid

• you just need simple row-based data handling with small datasets
• you need the full C++ feature set such as Flight SQL or Gandiva
• your project is not in Rust and native bindings exist for your language

Facets

library · maturity active

serialization parser data-science etl databases big-data analytics developer-tools rust cross-platform apache-arrow parquet columnar-format in-memory-analytics zero-copy dataframes data-engineering

Member repositories

For agents

markdown · JSON · MCP: product_card(name="apache/arrow-rs")