Configure CSV upload and storage options

Enable or disable CSV upload to allow or restrict users from writing CSV data to connected data platforms. When CSV upload is enabled, you can configure the following options to control the data flow between Sigma and your data platform:

• Internal or external stages (on Snowflake connections only): Choose to use internal or external stages to temporarily store CSV files before loading them to Snowflake.
• CSV upload with storage integration: Choose to use the default Sigma-owned cloud storage, or configure your organization to use customer-owned cloud storage for additional security and compliance.

This document explains how to set the organization-wide settings for CSV uploads. For information about how to upload CSV files, see Upload CSV data.

Requirements

The ability to configure CSV upload and storage options requires the following:

• You must be assigned the Admin account type.

• To enable customer-owned cloud storage for CSV upload, your organization must have an existing external storage integration that uses a customer-owned bucket.

For information about user requirements to upload CSV files, see Upload CSV data.

📘

CSV upload is only supported by connections with write access enabled. If the CSV Upload setting for your organization is turned on, but the connection a user wants to upload CSV data to does not have write access, the user will not be able to upload the CSV file.

Where uploaded CSV data is stored

When a user uploads a CSV file to Sigma, the data is written to your data platform as a new table in the schema used for write-back. If your organization uses multiple connections with write access enabled, users with access to more than one connection are required to choose one to write the CSV data to.

Before any data is written to your data platform, however, the data flow can follow one of two possible paths, depending on your organization settings. These paths differ in where the raw and processed CSV files are staged before the data is written to your data platform.

• Sigma-owned cloud storage (default): Staging files are temporarily stored in a Sigma-owned storage bucket in Sigma's infrastructure.
• Customer-owned cloud storage: Staging files are temporarily stored in a customer-owned storage bucket in your organization's cloud platform. See Choose Sigma-owned or customer-owned cloud storage for staging CSV files.

📘

If a user uploads a CSV file to a workbook, the data is only accessible from the specific workbook using the CSV file. When browsing the connection's data catalog, users cannot see the table created from the CSV upload, and if a user deletes the workbook, the data from the CSV file is no longer accessible.

To make CSV data available for reuse across multiple workbooks, upload the CSV file to a data model. You can also create a new data model from an existing workbook table or input table that was populated by a CSV upload.

Enable or disable CSV upload

Enable or disable the ability to upload CSV files. This setting only affects connections with write access enabled. Any connection that does not have write access cannot support CSV upload.

1. Open the Administrator portal.

2. In the side panel, select Account.

3. In the Features section, enable or disable the CSV Upload setting:

   • To enable CSV uploads, turn on the CSV Upload toggle.

     Users can immediately upload CSV files using connections with write access enabled.

   • To disable CSV uploads, turn off the CSV Upload toggle.

     Users are immediately prevented from uploading CSV files using any connection.

Choose internal or external stages for CSV uploads

📘

The option to use internal stages is only available for Snowflake connections.

If your organization is connected to Snowflake, you can choose to use internal or external stages to temporarily store CSV data before loading it to Snowflake. Internal stages are commonly used when the Snowflake account parameter REQUIRE_STORAGE_INTEGRATION_FOR_STAGE_CREATION must be set to false.

1. Open the Administrator portal.

2. In the side panel, select Account.

3. In the Features section, set the Internal stages setting (displayed only when the CSV Upload toggle is turned on) to determine how CSV files are temporarily stored before loading to Snowflake:

   • To use internal stages, turn on the Internal stages toggle.

   • To use external stages, turn off the Internal stages toggle.

4. Refer to the following Snowflake documentation for information about Snowflake's implementation of the temporary storage of CSV files in internal and external stages.

   • Bulk loading from a local file system (internal)

   • Bulk loading from Amazon S3 (external)

   • Bulk loading from Google Cloud Storage (external)

   • Bulk loading from Microsoft Azure (external)

Choose Sigma-owned or customer-owned cloud storage for staging CSV files (Beta)

By default, CSV upload uses a Sigma-owned storage bucket to temporarily store CSV staging files. Using Sigma-owned storage can be a simpler setup, but it may not meet your organization's security and compliance requirements. As an alternative, you can choose to use customer-owned cloud storage, which requires a storage integration that uses a customer-owned bucket.

For more information about the advantages of using a customer-owned bucket and how to configure the required storage integration, see Configure storage integration using customer-owned bucket.

1. Open the Administrator portal.

2. In the side panel, select Account.

3. In the Features section, set the CSV upload with storage integration setting to determine where staging files are stored:

   • To enable customer-owned cloud storage and store staging files in a customer-owned bucket, turn on the CSV upload with storage integration toggle.

   • To disable customer-owned cloud storage and store staging files in the default Sigma-owned bucket, turn off the CSV upload with storage integration toggle.

⚠️

When CSV upload with storage integration is enabled, but a storage integration isn't configured, CSV uploads will not work. When a user attempts to upload a CSV file, Sigma displays an error message indicating that no storage integration is configured.

For Snowflake connections, if both Internal stages and CSV upload with storage integration are enabled, Sigma uses internal stages and does not use the storage integration.