$5 free Limited time. Sign up to activate. Get your $5 →
Glossary · Feeds and delivery

What is Delta Sharing?

Delta Sharing is an open protocol for secure data sharing that lets a provider share live tables from a Delta Lake with recipients on other platforms, without copying the data. It is part of the Linux Foundation's Delta Lake project.

Also called: Delta Share · open data sharing protocol

Delta Sharing in practice

The provider exposes tables through a sharing server; recipients connect with a client (pandas, Apache Spark, Power BI and others) and read the data directly from the provider's cloud storage. The delta.io project page describes it as the industry's first open protocol for secure data sharing, announced at Data + AI Summit 2021.

For a B2B data buyer, Delta Sharing is the Databricks-world counterpart of a Snowflake share: the dataset appears in your lakehouse as tables you can query, governed and audited by the provider, with no files to download.

How Delta Sharing works

  1. Provider creates a shareTables in the provider's Delta Lake are added to a share and a recipient is granted access.
  2. Recipient gets a profileA credential file points the client at the sharing server.
  3. Client reads directlyThe client lists the shared tables and reads them from cloud storage (S3, ADLS, GCS) using short-lived URLs.
  4. Provider governsAccess can be tracked, audited and revoked at the share.
Delta Sharing: diagram Recipients read the shared tables directly from the provider's storage, on whatever platform they run. Provider DeltaLake Databricks Apache Spark pandas DeltaSharing
Recipients read the shared tables directly from the provider's storage, on whatever platform they run.

Delta Sharing vs Snowflake Secure Data Sharing

Delta SharingSnowflake share
ProtocolOpen, Linux Foundation projectSnowflake-native
Recipient platformAny client that speaks the protocolA Snowflake account (or a reader account)
Storage formatDelta Lake tables on cloud storageSnowflake tables
CopyingNone; read from provider storageNone; read through Snowflake's services layer

Limits and when not to use it

  • The provider has to run or host a sharing server and keep the shared tables in Delta format.
  • Reads go to the provider's cloud storage, so egress and region matter for large tables.
  • As with any share, what you may do with the data is set by the licence, not by the protocol.
With Forager

How Delta Sharing works with Forager

Forager delivers licensed feeds to Databricks customers as a data share, alongside Snowflake shares and file delivery to S3 and Google Cloud Storage. Tell us which lakehouse you run and we set up the delivery that fits.

  • $5 free credits
  • No credit card
  • Run your sample

Questions about Delta Sharing

What is Delta Sharing?

An open protocol for secure data sharing from the Delta Lake project. A provider shares live tables; recipients read them directly with clients such as pandas or Apache Spark, without the data being copied to another system.

Who maintains Delta Sharing?

It is part of the Delta Lake project, which the project describes as an independent open-source project under the Linux Foundation, and it was announced at Data + AI Summit 2021.

How is Delta Sharing different from a Snowflake share?

Both avoid copying. Delta Sharing is an open protocol that any client can speak; a Snowflake share works between Snowflake accounts through Snowflake's services layer.

Can I get Forager data into Databricks?

Yes. Forager delivers licensed feeds to Databricks as a data share, so the people, company and job posting tables appear in your lakehouse ready to join.

What does a Forager feed delivered by share cost?

Feeds are licensed by agreement and priced per record set, whichever delivery you choose. The delivery method does not change the price.

Start building

Test it on your own records

Create an account, start with $5 in credits and measure hit rate and connect rate on a sample you choose.

  • $5 free credits
  • No credit card
  • Run your sample
Platforms and teams that build on Forager
  • Clay
  • Juicebox
  • Nooks
  • Andela
  • HubSpot
  • SeekOut