How can you effectively use Snowflake for data sharing?

Snowflake data sharing is a great way to provide read-only access to your data in a secure way to internal and external consumers. It's very helpful in data monetization kind of scenarios and the consumers get to access the up to date data instead of offline files / feeds and can use the query / reporting tool of their choice to navigate the data. Consumers can have Snowflake account already or not. In case they do not, the Publisher of the data gets charged for the compute costs of data access / queries. It is important for publishers to be aware and put in the governance guardrails.

There are two main options for sharing data with Snowflake: Listing: You can offer a listing of your data as a product to one or more accounts, across regions and clouds. You can also provide additional metadata, charge for access, and offer your data publicly on the Snowflake Marketplace. You can view consumer usage metrics and automatically replicate your data to other regions. Direct Share: You can directly share specific database objects to another account in your region. This option is simpler and faster but does not support cross-region or cross-cloud sharing, charging, or metadata.

1. Provide read-only access to your data in a secure way to internal and external consumers. 2. Consumers do not require snowflake account. Client sharing the data will incur the expense. 3. Quick and easy to set up

Access control to data views are the best options. Other accounts or applications need to access create views on the data and share it

Data sharing is not limited to Snowflake accounts within the same organization. It can extend to external organizations, enabling collaboration and sharing across different entities.

CREATE SHARE: Used to create a new share. You specify the share name, the objects (e.g., databases, schemas, tables) you want to share, and the accounts with which you want to share. ALTER SHARE: Used to modify an existing share. You can use this command to make changes such as adding or removing objects, changing quotas, or adjusting permissions. DROP SHARE: Used to delete a share. This removes the share and its associated settings.

Creation of share is possible only by Accountadmin by default. However, the privilege 'Create Share' can be granted to other custom roles as well. Shares are created using CREATE SHARE <share name> SQL command. Once the share is created, the database, schema, and objects to be shared can be added to it. Managing the share could involve adding/dropping objects or consumer accounts from share. It can be done via ALTER SHARE <share name> command. Dropping of the share can be done by executing DROP SHARE <share name> command

Execute a CREATE DATABASE statement with the following data sharing-specific syntax: CREATE DATABASE <name> FROM SHARE <provider_account>.<share_name> Where provider_account is the name of the account that provided the share and share_name is the name of the share from which to create the database. A share can only be consumed once per account. To see the objects that are being shared before creating a database, use the DESCRIBE SHARE command. When a database is created from a share, only the role used to create the database can access objects in the database by default.

Best part about Snowflake data sharing via marketplace is how easy it is to set up, test if it fits our requirements using trials and the start consuming immediately. And later if we want to sell the enriched data in the marketplace, it’s again so easy to do so with data sharing.

Creation of share is possible only by Accountadmin by default. However, the privilege 'Import Share' can be granted to other custom roles as well. The shares can be imported via 2 methods - via Snowflake GUI - Executing SQL commands. The SQL command to consume shares is CREATE DATABASE <name> FROM SHARE <provider_account>.<share_name>

The platform provides dynamic access control, allowing data providers to adapt permissions and roles in real-time. This flexibility ensures that access is aligned with evolving data sharing agreements and business needs.

When sharing data to many accounts, it’s recommended to audit that those accounts are set up to consume that data and utilize that share. If you have accounts that aren’t using a configured data share and don’t plan to, you should limit your security footprint and remove those accounts from your share. To audit these data shares, below SQL commands that you can execute show shares; show grants to share myshare; show grants of share myshare;

Snowflake Data Sharing Options: a) Direct Share is a basic form of data sharing that includes the direct sharing of data between accounts. This includes three simple steps starting with creating a Share, adding objects to the share, and finally adding accounts. b) Snowflake Marketplace uses Secure Data Sharing to share data between data providers and consumers. It includes adding listings to Snowflake Marketplace post review and approval process and enables access to third-party data. c) Data Exchange refers to the user’s own data hub for data sharing between a few selected members.