Preview Features

Public Preview features

Public Preview features are intended for early testing and use by customers. They are stable, but can be limited in capability to allow exploration and testing. Within the documented limitations the feature is GA, and can be expected to evolve and change quickly with subsequent releases.

Feature	Release	Description	Additional Information
Partition-Based Incremental Aggregates	C2026.2.0	Partition-based incremental aggregates enable you to leverage your data warehouse's native partition (or cluster block) pruning capabilities when building and querying incremental aggregates. This can help reduce run-time query overhead observed in the view-based implementation of incremental aggregates.	About Incremental Rebuilds, Setting Properties to Allow Incremental Rebuilds of Aggregates
Slowly Changing Dimensions	C2025.12.0	Slowly changing dimensions (SCDs) are used in business intelligence to track how dimension data (like a customer’s address or a product's category) changes over time. Instead of overwriting old values, SCDs keep historical records. This allows analysts to see what data was available at the time of the transaction, which is critical for audit trails, machine learning, and accurate reporting.	Working with Slowly Changing Dimensions
AtScale MCP Server	C2025.11.0	AtScale provides its own MCP Server, enabling you to connect to and query your AtScale data using AI applications like ChatGPT and Claude.	Connecting to AtScale with AI
Constraint Translation	C2025.5.0	Constraint translation enables the AtScale engine to translate dimension filter constraints in queries to fact table partition column constraints. When the data warehouse processes the query, it can use the additional partition column constraints to accelerate it.	Using Constraint Translation for Fact Table Query Acceleration
XIRR MDX function	C2025.2.0	The XIRR MDX function calculates the rate of return (within .000001%) of investments that have irregular payment schedules. Note that this is only supported for the following data warehouses: Google BigQuery, Redshift, Snowflake, and PostgreSQL.	XIRR

Private Preview features

Private Preview features are limited to selected customers for feedback gathering. They are scoped and tested for specific initial use cases. The intention is to move these features into Public Preview once sufficient feedback has been gathered and a broader use case portfolio has been collected. Major and minor changes to the feature should be expected during this stage.

If you wish to be included during Private Preview, reach out to your account representatives.

Note

Public documentation for Private Preview features will be made available once they move into Public Preview.

Feature	Release	Description
Preferred Aggregate Storage	C2026.2.0	Preferred Aggregate Storage (PAS) offers an aggregate storage option for the aggregate system. When enabled, the system promotes aggregates that fall under configured settings from the data platform to the preferred aggregate store. By default, the preferred aggregate store is a container on the Kubernetes deployment.