zia#

A benchmarking framework for evaluating compression algorithms on numeric array datasets, with a focus on scientific data.

Overview#

Zia provides systematic benchmarking of compression methods for numeric arrays, measuring:

• Compression ratio
• Encoding throughput (MB/s)
• Decoding throughput (MB/s)

Repository Structure#

Key components are located in the zia_benchmark/src/zia_benchmark/ directory:

• Algorithms: /algorithms/

  • LZMA: /algorithms/lzma/
  • Zstandard: /algorithms/zstd/
  • ZLIB: /algorithms/zlib/
  • Simple ANS: /algorithms/simple_ans/
  • etc.

• Datasets: /datasets/

  • Synthetic data generators:
    • /datasets/bernoulli/: Binary random data
    • /datasets/gaussian/: Normal distribution samples
  • Real data: /datasets/real/
  • etc.

• Core functionality:

  • run_benchmarks.py: Main benchmarking engine

Development#

Code Formatting#

This project uses pre-commit hooks to automatically format code before each commit. The formatting includes:

• Python code formatting using black
• TypeScript/JavaScript code formatting using npm scripts

To set up the pre-commit hooks after cloning the repository:

1. Install pre-commit:

pip install pre-commit

2. Install the git hook scripts:

pre-commit install

After this setup, code will be automatically formatted when you make a commit.

Results#

Latest benchmark results: https://magland.github.io/zia/