Skip to main content

Btune

Intelligent data compression

Btune helps you find better compression parameters automatically, so you can optimize for the metrics that matter most in your workload.

Btune optimization diagram

What it is

Compression tuning without exhaustive trial and error

Effective compression depends on data type, structure, use case, and changing operating conditions. Btune automates parameter selection so you do not have to search the space manually.

Btune Community

A free edition for personal use that explores parameter combinations with a genetic algorithm.

Btune Models

A commercial license with a custom neural model trained on your sample data for fast recommendations.

Btune Studio

A full commercial solution for training and deploying your own models on-site across many datasets.

Why it matters

Optimize for speed, ratio, or the right trade-off

Btune helps you systematically choose the balance between compression ratio, compression speed, and decompression speed that best fits your application.

Compression ratio versus compression speedCompression ratio versus decompression speedCompression ratio versus codec

Btune finds strong parameter combinations in the Pareto sense, so you can choose the best balance for your own requirements.

How to use it

Start quickly, then scale up

Install from PyPI

The same plugin supports Btune Community and Btune Models.

pip install blosc2-btune

For details, see the Btune README. Btune Studio requires additional on-site training software; contact us for setup.

What's in a model?

Train once, predict instantly

Simple neural network model

Btune stores trained models as small files and uses them to predict the best compression parameters for each chunk of data in real time.

Example

Optimization at scale

In one example, Btune was used to optimize decompression speed for a 7.3 TB subset of the Gaia dataset while balancing file size and access speed.

Predicted codecs and filters for decompressionSlicing speed for different codecs and filtersFile sizes for different codecs and filters

For more detail, see our SciPy 2023 paper, slides, and the example repository.

Testimonials

What users say

"Blosc2 and Btune are fantastic tools that allow us to efficiently compress and load large volumes of data for the development of AI algorithms for clinical applications."

Leonardo Ayala, Div. Intelligent Medical Systems, German Cancer Research Center

"Btune is a simple and highly effective tool. We tried this with LEAPS data and found some super useful spots in the parameter space of Blosc2 compression arguments."

Peter Steinbach, Helmholtz AI Consultants Team Lead for Matter Research

Pricing

Ready to optimize your compression strategy?

Compare licensing options or contact us to discuss Btune for your workload.