Feature Request for Apache Kylin: Implementing Row-Level and Partition-Level Security

[DataSymphony]

It would be great if Apache Kylin could be enhanced to implement of row-level and partition-level security features, similar to those found in Microsoft Analysis Services.

By providing row-level and partition-level security, Apache Kylin could offer finer-grained access control, allowing enterprises to enforce their security policies more effectively. This is crucial for organizations that have to meet strict compliance requirements or deal with sensitive data.

Potential Requirements

1. Row-Level Security

Objective

To give administrators the ability to restrict user access to specific rows of data within a cube based on cube dimensions.

Features

• Integration with Active Directory (AD) or alternative user group management systems to authenticate and authorize users.
• Ability to specify access rules based on cube dimensions.
• Rule-based expressions to match specific conditions within rows.
• Easy-to-use management interface for administrators to manage rules and user permissions.

Use Cases

• Departments can restrict access to salary or performance evaluation data for different user groups.
• Financial data can be restricted so that only the relevant departmental users have access.

2. Partition-Level Security

Objective

To give administrators the ability to restrict user access to specific partitions within a cube.

Features

• Integration with Active Directory (AD) or alternative user group management systems.
• Ability to specify which partitions a user or group can access.
• Easy-to-use management interface for administrators to manage partition-level permissions.

Use Cases

• Partitioning data by business segment and then allowing users to only access data relevant to their segment.
• Time-based partitioning where access is restricted based on date ranges for historical or forecasting data.