Multitenant Database View Aggregation Without Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multitenant database systems, managing unified views across multiple tenants corresponding to the same entity is challenging, as existing solutions often require replicating data to external systems, which is costly and resource-intensive, and do not provide real-time data access without accessing each separate instance.
Innovation Solution
A new database instance is provisioned for the common entity, with connectors configured to access data from separate tenant instances without replication, using foreign data wrappers to maintain data locality and generate a unified view, allowing a new cloud platform instance to access this data without significant reconfiguration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If data is replicated to external systems to provide unified views, then data accessibility is improved, but resource overhead and cost increase
Solution Approach 1:
The patent introduces a virtualization layer that acts as an intermediary between the multitenant database system and the data access interface. This virtualization layer provides unified views of data from multiple tenants without physically replicating the data, thereby maintaining data locality while enabling efficient access. The virtualization layer abstracts the complexity of data aggregation and allows users to access consolidated data across tenants without the resource-intensive process of data replication.
2Loss of information
If separate instances are accessed to obtain complete data, then data completeness is improved, but access complexity increases
Solution Approach 1:
The patent merges multiple tenant-specific data views into a single unified view through virtualization. Instead of requiring users to access separate database instances for each tenant, the system combines data from multiple tenants into one consolidated interface. This merging process maintains the completeness of data while significantly reducing access complexity, as users interact with a single unified system rather than multiple separate instances.
Solution Approach 2:
The virtualization layer provides a universal interface that can access and present data from multiple tenant instances through a single point of entry. This multi-functional interface handles data aggregation, filtering, and presentation across different tenants, eliminating the need for separate access procedures for each tenant and simplifying the overall access process.
3Adaptability or versatility
If data is replicated externally, then unified view capability is improved, but data locality is lost
Solution Approach 1:
The virtualization layer serves as an intermediary that enables unified view capability without moving data from its original location. It creates virtual representations of data from multiple tenants while the actual data remains in place in the respective tenant database instances. This approach maintains data locality and associated performance benefits while still providing the versatility of unified data access and analysis.
Data Source
AI summary
A server may support an aggregation service for a multitenant system. The service may support a method for data processing that includes determining that a plurality of tenants of a multitenant database system correspond to a common entity, where each tenant is associated with a respective instance of a cloud platform that services data from a corresponding database instance. The method may further include provisioning a new database instance for the common entity, generating a connector that is configured to access data of the corresponding database instance, generating a schema for the new database instance that aggregates the data of each database instance accessed by the connector, and instantiating a new cloud platform instance that is configured to access the schema of the new database instance, where the new cloud platform instance is configured to serve data queried from the schema of the new database instance via a client application.