Zed Tests

The Zed system provides an open-source, cloud-native, and searchable data lake for semi-structured and structured data.

Zed lakes utilize a superset of the relational and JSON document data models yet require no up-front schema definitions to insert data. They also provide transactional views and time travel by leveraging a git-like design pattern based on a commit journal. Using this mechanism, a lake’s (optional) search indexes are transactionally consistent with its data.

At Zed’s foundation lies a new family of self-describing data formats based on the Zed data model, which unifies the highly structured approach of dataframes and relational tables with the loosely structured document model of JSON.

While the Zed system is built around its family of data formats, it is also interoperable with popular data formats like CSV, (ND)JSON, and Parquet.

This repository contains tools and components used to organize, search, analyze, and store Zed data, including:

  • The zed command line tool for managing, searching, and querying a Zed lake
  • The Zed language documentation
  • The Zed formats specifications and documentation

The previously released zq tool is now packaged as a command-line shortcut for the zed query command.


To install zed or any other tool from this repo, you can either clone the repo and compile from source, or use a pre-compiled release, available for Windows, macOS, and Linux.

If you don’t have Go installed, download and install it from the Go downloads page. Go version 1.16 or later is required.

To install the binaries in $GOPATH/bin, clone this repo and execute make install:

git clone https://github.com/brimdata/zed
cd zed
make install


See the contributing guide on how you can help improve Zed!

Join the Community

Join our Public Slack workspace for announcements, Q&A, and to trade tips!