What if you could manage your data lake just like you manage code? With rollback, versioning, and branching capabilities on top of your existing data lake?
lakeFS is an open-source project that provides a Git-like version control interface for data lakes, with seamless integration to most data tools and frameworks. lakeFS enables you to easily implement parallel pipelines for experimentation, reproducibility, and CI/CD for data.
And now you can start playing around with lakeFS in a fully functional lakeFS environment, with your own data and all the tools that you are already using. Get your isolated environment, integrate it with the tools you use, and see how it works in an environment similar to your own.Â
Please note: playground environments will be deleted after one week.
See these features in action:
- Full reproducibility of data and code
- Git-like operations: branch, commit, merge and revert
- Instant reversion of changes to data
- Petabytes scale version control
- Zero-copy branching for frictionless experimentation
- Seamless integration with most data tools and frameworks (Spark, Hive, AWS Athena, Presto)
- Support for AWS S3, Azure Blob Storage, and Google Cloud Storage (GCS)
Start playing here (no email or any additional information required)
One you have experienced the value in lakeFS, go ahead and install lakeFS open source, or sign up for lakeFS Cloud Beta. And don’t forget to share your feedback with us via our community on Slack or GitHub – this is the best way to let us know if any feature you need is missing!Read Related Articles.


What is Metadata? Ultimate Guide For Data Engineers
What is metadata? Why is it so important? Keep reading to learn more about modern practices in metadata management.


The Power of Databricks SQL: A Practical Guide to Unified Data Analytics
In the universe of Databricks Lakehouse, Databricks SQL serves as a handy tool for querying and analyzing data. It lets SQL-savvy data analysts, data engineers,


ML Data Version Control and Reproducibility at Scale
Introduction In the ever-evolving landscape of machine learning (ML), data stands as the cornerstone upon which triumphant models are built. However, as ML projects expand
Table of Contents