https://github.com/maxi-k/btrblocks
BtrBlocks: Efficient Columnar Compression for Data Lakes (SIGMOD 2023 Paper)
https://github.com/maxi-k/btrblocks
compression data-lake databases research
Last synced: 2 months ago
JSON representation
BtrBlocks: Efficient Columnar Compression for Data Lakes (SIGMOD 2023 Paper)
- Host: GitHub
- URL: https://github.com/maxi-k/btrblocks
- Owner: maxi-k
- License: mit
- Created: 2023-05-22T15:15:22.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2025-04-07T10:12:31.000Z (2 months ago)
- Last Synced: 2025-04-09T15:08:02.976Z (2 months ago)
- Topics: compression, data-lake, databases, research
- Language: C++
- Homepage:
- Size: 247 KB
- Stars: 238
- Watchers: 7
- Forks: 20
- Open Issues: 8
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# BtrBlocks - Efficient Columnar Compression for Data Lakes
[](https://bit.ly/btrblocks)
[](https://github.com/maxi-k/btrblocks/actions/workflows/test.yml)- [Paper](https://bit.ly/btrblocks) (two-column version)
- [Video](https://dl.acm.org/doi/10.1145/3589263) (SIGMOD 2023 presentation)## Usage
After [building](#building) the library, follow one of the [examples](./tools/examples) to get started.
## Components
- `btrblocks/`: the compression library, schemes, utilities, ...
- `btrfiles/`: helper library for binary files and yaml schema information
- `tools/`: various conversion, measurement and benchmarking tools
- `test/`: rudimentary tests for the library
## Building
We currently require an x86 platform.
This library was built and tested on linux only.``` sh
mkdir build
cd build
cmake ..
```Then, depending on your usecase, build only the library or any of the tools:
- build everything: `make`
- install static library and headers on your system: `sudo make install`
- build the compression library only: `make btrblocks`
- build the tests `make tester`
- build the in-memory decompression speed benchmark: `make decompression_speed`
- ...For a list of all valid targets, run `make help`.
Library was built and tested on Linux (x86, ARM) and MacOS (ARM).
## Contributors
Adnan Alhomssi
David Sauerwein
Maximilian Kuschewski## License
MIT - See [License File](LICENSE)