Custom Multi-Tenant Objects in Non-Relational Databases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-relational database systems, such as HBase, face challenges in dynamically updating schema elements and provisioning access rights without interrupting real-time data collection, as traditional methods require locking down the database and causing downtime.
Innovation Solution
Implementing custom base platform objects (BPOs) defined in scripting languages like XML, which allow for dynamic virtual tables with shared schemas, enabling updates and access control at the application layer without physical changes to the database structure, thus allowing zero-downtime schema management and personalized access for tenants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional schema update methods are used in non-relational database systems, then schema changes can be made, but the database must be locked down causing downtime and interrupting real-time data collection
Solution Approach 1:
The patent segments the schema management into two independent layers: a metadata layer for schema definitions and a data layer for actual operations. This allows schema updates to occur in the metadata layer without affecting the data layer, enabling continuous real-time data collection while schema changes are applied.
Solution Approach 2:
The patent introduces a virtual table layer as an intermediary between the physical non-relational database and the application layer. This virtual table handles schema updates and translates them to the physical database, allowing schema changes without requiring the database to be locked down, thus maintaining continuous data collection operations.
2Adaptability or versatility
If physical changes are made to the database structure for schema updates, then schema evolution is achieved, but real-time data operations are disrupted causing downtime
Solution Approach 1:
The patent performs schema validation and compilation in advance before applying changes to the physical database. The virtual table pre-processes schema updates, validating them against the non-relational database constraints beforehand, so that when changes are applied, they can be executed quickly without prolonged downtime.
Solution Approach 2:
The patent implements dynamic schema management where the virtual table can adapt schema changes in real-time without requiring static physical database restructuring. Schema evolution occurs dynamically through metadata updates and virtual table reconfiguration, allowing continuous operations without the downtime associated with static physical changes.
3Reliability
If access control is implemented at the database level, then security is improved, but flexibility for personalized tenant access is reduced
Solution Approach 1:
The patent implements access control with local quality by allowing different access policies to be applied to different tenants and data objects independently. The virtual table enables fine-grained, personalized access control rules for each tenant while maintaining overall database security, rather than applying uniform access control at the database level.
Solution Approach 2:
The patent adds a new dimension to access control by implementing a virtual table layer between the physical database and users. This additional layer enables personalized access control policies to be applied without compromising the underlying database security model, allowing flexible tenant-specific permissions while maintaining robust database-level security.
4Adaptability or versatility
If custom objects are added to extend functionality for specific tenants, then adaptability is improved, but database complexity increases
Solution Approach 1:
The patent uses copying by creating virtual representations of custom objects in the virtual table layer rather than physically altering the database structure for each custom object. Tenant-specific functionality is achieved by copying and adapting schema definitions in the metadata layer, allowing custom objects to be added without increasing physical database complexity.
Data Source
AI summary
Disclosed are examples of systems, apparatus, methods and computer program products for creating custom platform objects for a multi-tenant non-relational database environment. A database system maintains a multi-tenant non-relational database associated with multiple enterprises and a number of records for each of the enterprises. The system also maintains a dynamic virtual table associated with a number of records. A request is received from one of the enterprises to define a custom data object within the database system, uniquely associated with an enterprise. The request is processed, and a custom object is generated based on the request. The custom object defines database columns corresponding to the data object and the enterprise. The virtual table is updated to include virtual columns corresponding to the database column definitions, and existing columns of a shared table in the non-relational database are updated to match the virtual columns. Access to the existing columns is then restricted for enterprises not uniquely associated with the custom data object.