Why a Snowflake Native Application?

Casabase Cube is built and distributed using the Snowflake Native App Framework. Rather than operating as a separate SaaS analytical platform or external OLAP server, Casabase Cube is installed and runs within the customer’s Snowflake environment.

This architecture allows organizations to add multidimensional OLAP capabilities to Snowflake while continuing to use Snowflake as the platform for storage, compute, identity, security, and governance.

Keep Cube Processing in Snowflake

Traditional OLAP architectures commonly require data to be moved from a source platform into a separate cube server or analytical database, where additional aggregate structures may then be built and maintained.

Casabase Cube removes the separate OLAP platform from that architecture.

Data loaded into a Casabase Cube is stored in application-managed Snowflake tables within the Casabase Cube Native Application database. Hierarchy processing, multidimensional aggregation, formula evaluation, security enforcement, and query execution occur within the customer’s Snowflake environment.

This provides several advantages:

• No separate OLAP server or analytical database
• No Casabase Software-hosted data platform required for cube processing
• No separate precomputed aggregate store to build and maintain
• Snowflake remains the data and processing boundary for Casabase Cube
• Existing Snowflake identity, governance, and security controls remain part of the overall architecture

See Data Storage & Boundaries for details.

Snowflake-Native Access Control

Because Casabase Cube is a Snowflake Native App, access to application functionality is integrated with Snowflake.

Casabase Cube defines application roles:

• CUBE_ADMIN provides administrative access and inherits CUBE_PUBLIC.
• CUBE_PUBLIC provides access to client-facing Casabase Cube functionality, including supported querying and shared views.

Snowflake administrators grant these application roles to the appropriate Snowflake account roles.

Casabase Cube does not require customers to maintain a separate Casabase Cube authentication system.

Application access is separate from Casabase Cube’s optional user-based row-level security. A user can have permission to use Casabase Cube while still being restricted to particular portions of a cube’s hierarchies.

See Security Model for details.

No Separate OLAP Infrastructure

Traditional OLAP platforms typically require organizations to deploy and maintain dedicated infrastructure for the analytical engine.

With Casabase Cube, customers do not provision a separate Casabase Cube OLAP server, virtual machine, container platform, or analytical database.

Snowflake provides the underlying platform for:

• Compute
• Storage
• Authentication
• Authorization
• Governance
• Scalability
• Native App deployment

Casabase Cube provides the multidimensional engine and application-managed cube model within that platform.

Dynamic Aggregation

Casabase Cube does not require stored parent-level aggregates to be built before users can query the cube.

Instead:

Bottom-Level Cube Data
        +
Hierarchy Metadata
        +
Aggregation Rules
        +
Formulas
        +
Security
        │
        ▼
Query-Time Processing
        │
        ▼
Multidimensional Result

Parent values and calculated members are evaluated dynamically when required by a query.

This removes the traditional cube-processing cycle used to build and refresh aggregate structures.

Use Snowflake Compute

Casabase Cube uses Snowflake compute for analytical and application workloads, including:

• Multidimensional queries
• Dynamic aggregation
• Formula evaluation
• Hierarchy processing
• Data imports
• Cube administration and maintenance

This allows Casabase Cube workloads to use Snowflake’s compute model rather than requiring a separate compute tier dedicated to the OLAP engine.

Query performance still depends on factors such as warehouse capacity, data volume, query grain, hierarchy expansion, and calculation complexity.

Customer-Controlled External Access

Some Casabase Cube operations require access to customer-owned Snowflake objects outside the Native Application database.

Examples can include:

• Snowflake stages used for Essbase or Cloud EPM migration
• Customer tables used as sources for cube or dimension creation
• Other explicitly supported integration objects

Access to these objects must be explicitly granted to the application.

Granting access to one approved object does not give Casabase Cube unrestricted access to unrelated customer databases, schemas, tables, or stages.

Simplified Deployment and Updates

Casabase Cube is distributed through Snowflake, including Snowflake Marketplace distribution.

This provides a controlled deployment model within the customer’s Snowflake environment and avoids installing a separate OLAP software stack.

Application updates are delivered through Snowflake’s Native App distribution model.

For operational considerations associated with upgrades, see Application Upgrades.

Integrated with the Snowflake Ecosystem

Casabase Cube exposes client-facing functionality through Snowflake stored procedures and supported relational views.

Applications can therefore integrate using standard Snowflake connectivity, including:

• Snowflake SQL
• Snowflake APIs
• Snowflake-supported drivers and connectors
• Business intelligence and visualization platforms
• Custom applications
• Data and analytics workflows

Casabase Cube does not require a separate proprietary external API or network service for cube queries.

See Integration Points for details.

Snowflake Intelligence

Casabase Cube can also participate in Snowflake Intelligence and Cortex Agent workflows.

The conversational layer can interpret a user’s request and invoke configured Casabase Cube capabilities, while the Casabase Cube engine remains responsible for governed multidimensional query behavior.

Snowflake Intelligence adds a natural-language interface. It does not replace Casabase Cube’s hierarchy, formula, aggregation, or security semantics.

See Intelligence Overview.

A Modern Alternative to Traditional OLAP Architecture

At a high level:

Traditional OLAP

Source Data
    │
    ▼
Separate OLAP Platform
    │
    ├── Cube Storage
    ├── Aggregate Processing
    └── Calculation Engine
    │
    ▼
Analytical Tools

Casabase Cube

Customer's Snowflake Account
    │
    ├── Source Data
    │
    └── Casabase Cube Native App
            ├── Application-Managed Cube Data
            ├── Hierarchies and Metadata
            └── Query-Time Calculation Engine
                    │
                    ▼
             Analytical Tools and AI

The key architectural difference is that the multidimensional engine, its managed cube data, and its analytical processing operate within Snowflake without a separate OLAP platform or precomputed aggregate store.

Next Steps

• Review Key Terminology to become familiar with Casabase Cube concepts.
• Follow the 5-Minute Quick Start to begin using Casabase Cube.
• Explore High-Level Architecture for a deeper look at how Casabase Cube works.