Skip to main content

Open Protocol for Secure Data Sharing

Share large datasets securely in real-time across organizations without copying data. Delta Sharing is an open REST protocol that works with any computing platform.

Get StartedView Protocol
import
delta_sharing

client = delta_sharing.SharingClient(profile)
df = delta_sharing.load_as_pandas(table_url)

Quick Start

Get started with Delta Sharing in minutes using Python or Apache Spark.

1

Install the Python Connector

Install the Delta Sharing Python package using pip:
pip install delta-sharing
Requires Python 3.8+ and glibc >= 2.31 for automatic installation. See the installation guide for alternative methods.
2

Get a Profile File

A profile file contains credentials to access a Delta Sharing server. Download an example profile to try with our sample data:
curl https://databricks-datasets-oregon.s3-us-west-2.amazonaws.com/delta-sharing/share/open-datasets.share -o config.share
Profile files are JSON documents containing server endpoint and authentication details:
{
  "shareCredentialsVersion": 1,
  "endpoint": "https://sharing.delta.io/delta-sharing/",
  "bearerToken": "your-token-here"
}
3

Load Shared Data

Use the Python connector to load shared tables as pandas DataFrames:
import delta_sharing

# Create a SharingClient
client = delta_sharing.SharingClient("config.share")

# List available tables
tables = client.list_all_tables()

# Load a table as a pandas DataFrame
table_url = "config.share#share.schema.table"
df = delta_sharing.load_as_pandas(table_url)
If running in PySpark, use load_as_spark() to load tables as Spark DataFrames for distributed processing.

Explore by Topic

Choose your path based on how you want to use Delta Sharing.

Python Connector

Read shared tables with pandas or PySpark. Perfect for data science and analytics workflows.

Apache Spark Connector

Access shared data in Spark SQL, Python, Scala, Java, or R with full streaming support.

Reference Server

Deploy your own Delta Sharing server to share Delta Lake and Parquet tables.

Protocol Specification

Learn the open REST protocol that powers Delta Sharing integrations.

Key Features

Everything you need for secure, scalable data sharing.

Secure by Design

Bearer token authentication with pre-signed URLs for direct cloud storage access. No data passes through the server.

Multi-Cloud Support

Works with S3, Azure Blob Storage, Azure Data Lake Gen2, Google Cloud Storage, and Cloudflare R2.

Change Data Feed

Query incremental changes between versions for efficient incremental data processing and streaming.

Open Protocol

Simple REST protocol that anyone can implement. Community connectors available for 10+ languages and platforms.

Resources

Additional resources to help you succeed with Delta Sharing.

Community Connectors

Explore connectors for Rust, Go, Java, R, Node.js, and more.

Building from Source

Compile and test the Python connector, Spark connector, and server.

Contributing

Learn how to contribute to the Delta Sharing project.

Ready to start sharing data?

Follow our quickstart guide to share or consume your first Delta table in minutes.

Start Quickstart →