base on An extensible, state of the art columnar file format. Formerly at @spiraldb, now an Incubation Stage project at LFAI&Data, part of the Linux Foundation. # 🌪️ Vortex [](https://github.com/vortex-data/vortex/actions) [](https://www.bestpractices.dev/projects/10567) [](https://docs.vortex.dev) [](https://codspeed.io/vortex-data/vortex) [](https://crates.io/crates/vortex) [](https://pypi.org/project/vortex-data/) [](https://central.sonatype.com/artifact/dev.vortex/vortex-spark) [](https://codecov.io/github/vortex-data/vortex) [Join the community on Slack!](https://vortex.dev/slack) | [Documentation](https://docs.vortex.dev/) | [Performance Benchmarks](https://bench.vortex.dev) ## Overview Vortex is a next-generation columnar file format and toolkit designed for high-performance data processing. It is the fastest and most extensible format for building data systems backed by object storage. It provides: - **Blazing Fast Performance** - 100x faster random access reads (vs. modern Apache Parquet) - 10-20x faster scans - 5x faster writes - Similar compression ratios - Efficient support for wide tables with zero-copy/zero-parse metadata - **Extensible Architecture** - Modeled after Apache DataFusion's extensible approach - Pluggable encoding system, type system, compression strategy, & layout strategy - Zero-copy compatibility with Apache Arrow - **Open Source, Neutral Governance** - A Linux Foundation (LF AI & Data) Project - Apache-2.0 Licensed - **Integrations** - Arrow, DataFusion, DuckDB, Spark, Pandas, Polars, & more - Apache Iceberg (coming soon) > 🟢 **Development Status**: Library APIs may change from version to version, but we now consider > the file format <ins>_stable_</ins>. From release 0.36.0, all future releases of Vortex should > maintain backwards compatibility of the file format (i.e., be able to read files written by > any earlier version >= 0.36.0). ## Key Features ### Core Capabilities - **Logical Types** - Clean separation between logical schema and physical layout - **Zero-Copy Arrow Integration** - Seamless conversion to/from Apache Arrow arrays - **Extensible Encodings** - Pluggable physical layouts with built-in optimizations - **Cascading Compression** - Support for nested encoding schemes - **High-Performance Computing** - Optimized compute kernels for encoded data - **Rich Statistics** - Lazy-loaded summary statistics for optimization ### Technical Architecture #### Logical vs Physical Design Vortex strictly separates logical and physical concerns: - **Logical Layer**: Defines data types and schema - **Physical Layer**: Handles encoding and storage implementation - **Built-in Encodings**: Compatible with Apache Arrow's memory format - **Extension Encodings**: Optimized compression schemes (RLE, dictionary, etc.) ## Quick Start ### Installation #### Rust Crate All features are exported through the main `vortex` crate. ```bash cargo add vortex ``` #### Python Package ```bash uv add vortex-data ``` #### Command Line UI (vx) For browsing the structure of Vortex files, you can use the `vx` command-line tool. ```bash # Install latest release cargo install vortex-tui --locked # Or build from source cargo install --path vortex-tui --locked # Usage vx browse <file> ``` ### Development Setup #### Prerequisites (macOS) ```bash # Optional but recommended dependencies brew install flatbuffers protobuf # For .fbs and .proto files brew install duckdb # For benchmarks # Install Rust toolchain curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh # or brew install rustup # Initialize submodules git submodule update --init --recursive # Setup dependencies with uv uv sync --all-packages ``` ### Benchmarking Use `vx-bench` to run benchmarks comparing engines (DataFusion, DuckDB) and formats (Parquet, Vortex): ```bash # Install the benchmark orchestrator uv tool install "bench_orchestrator @ ./bench-orchestrator/" # Run TPC-H benchmarks vx-bench run tpch --engine datafusion,duckdb --format parquet,vortex # Compare results vx-bench compare --run latest ``` See [bench-orchestrator/README.md](bench-orchestrator/README.md) for full documentation. ### Performance Optimization For optimal performance, we suggest using [MiMalloc](https://github.com/microsoft/mimalloc): ```rust,ignore #[global_allocator] static GLOBAL_ALLOC: MiMalloc = MiMalloc; ``` ## Project Information ### License Licensed under the Apache License, Version 2.0. ### Governance Vortex is an independent open-source project and not controlled by any single company. The Vortex Project is a sub-project of the Linux Foundation Projects. The governance model is documented in [CONTRIBUTING.md](CONTRIBUTING.md) and is subject to the terms of the [Technical Charter](https://vortex.dev/charter.pdf). ### Contributing Please **do** read [CONTRIBUTING.md](CONTRIBUTING.md) before you contribute. ### Reporting Vulnerabilities If you discover a security vulnerability, please email <[email protected]>. ### Trademarks Copyright © Vortex a Series of LF Projects, LLC. For terms of use, trademark policy, and other project policies please see <https://lfprojects.org> ## Acknowledgments The Vortex project benefits enormously from groundbreaking work from the academic & open-source communities. ### Research in Vortex - [BtrBlocks](https://www.cs.cit.tum.de/fileadmin/w00cfj/dis/papers/btrblocks.pdf) - Efficient columnar compression - [FastLanes](https://www.vldb.org/pvldb/vol16/p2132-afroozeh.pdf) & [FastLanes on GPU](https://dbdbd2023.ugent.be/abstracts/felius_fastlanes.pdf) - High-performance integer compression - [FSST](https://www.vldb.org/pvldb/vol13/p2649-boncz.pdf) - Fast random access string compression - [ALP](https://ir.cwi.nl/pub/33334/33334.pdf) & [G-ALP](https://dl.acm.org/doi/pdf/10.1145/3736227.3736242) - Adaptive lossless floating-point compression - [Procella](https://dl.acm.org/citation.cfm?id=3360438) - YouTube's unified data system - [Anyblob](https://www.durner.dev/app/media/papers/anyblob-vldb23.pdf) - High-performance access to object storage - [ClickHouse](https://www.vldb.org/pvldb/vol17/p3731-schulze.pdf) - Fast analytics for everyone - [MonetDB/X100](https://www.cidrdb.org/cidr2005/papers/P19.pdf) - Hyper-Pipelining Query Execution - [Morsel-Driven Parallelism](https://db.in.tum.de/~leis/papers/morsels.pdf): A NUMA-Aware Query Evaluation Format for the Many-Core Age - [The FastLanes File Format](https://github.com/cwida/FastLanes/blob/dev/docs/specification.pdf) - Expression Operators ### Vortex in Research - [Anyblox](https://gienieczko.com/anyblox-paper) - A Framework for Self-Decoding Datasets - [F3](https://dl.acm.org/doi/pdf/10.1145/3749163) - Open-Source Data File Format for the Future ### Open Source Inspiration - [Apache Arrow](https://arrow.apache.org) - [Apache DataFusion](https://github.com/apache/datafusion) - [parquet2](https://github.com/jorgecarleitao/parquet2) by Jorge Leitao - [DuckDB](https://github.com/duckdb/duckdb) - [Velox](https://github.com/facebookincubator/velox) & [Nimble](https://github.com/facebookincubator/nimble) #### Thanks to all contributors who have shared their knowledge and code with the community! 🚀 ", Assign "at most 3 tags" to the expected json: {"id":"11982","tags":[]} "only from the tags list I provide: []" returns me the "expected json"