Intermediary Server Logical Table Partitioning for Cloud Database Multitenancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud database systems sacrifice flexibility and access control for performance, with expensive and inflexible table creation mechanisms, making it difficult to manage data separation and access rights across multiple applications.
Innovation Solution
The method of partitioning native tables into logical tables using an intermediary server that translates client queries, allowing clients to interact with logical tables as if they were full-fledged database tables, while maintaining data separation and access control without modifying existing clients or database clusters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single database table is created across all machines in the cluster, then performance and throughput are improved, but flexibility and access control are worsened
Solution Approach 1:
The patent segments the single database table into multiple logical tables, each serving specific applications or data types. This segmentation allows different applications to access only their designated portions of the data, improving flexibility and access control while maintaining the underlying single-table performance benefits of the cloud database cluster.
2Adaptability or versatility
If multiple tables are allowed in cloud databases, then flexibility is improved, but table creation becomes expensive and inflexible
Solution Approach 1:
The patent creates lightweight logical table definitions that reference the underlying native table structure. Instead of creating expensive physical table copies across the cluster, the system uses virtual table definitions with mapping relationships, allowing rapid table creation without the overhead of full table provisioning on each cluster node.
3Adaptability or versatility
If new tables are created in cloud databases, then data separation is improved, but coordination complexity and data corruption risks increase
Solution Approach 1:
The patent introduces an intermediary layer between applications and the underlying native tables. This intermediary manages the mapping between logical tables and physical storage, handling coordination and access control automatically. Applications interact with logical tables through defined interfaces, eliminating the need for direct coordination between applications and reducing data corruption risks.
4Ease of manufacture
If fixed table layouts and column types are enforced, then manufacturing ease is improved, but adaptability for arbitrary data storage is worsened
Solution Approach 1:
The patent implements dynamic table schemas where column types and layouts can be modified without recreating the entire table structure. The logical table definitions allow for flexible schema evolution, enabling arbitrary data storage while maintaining the ease of table creation through standardized logical table templates that adapt to different data requirements.
Data Source
AI summary
Methods and apparatus are described for partitioning native tables in a database cluster into logical tables. Each logical table is mapped into a unique portion of the native table by an intermediary server. Clients access a logical table as an ordinary, full-fledged database table through the intermediary server, which translates queries on the logical table into queries on the corresponding portion of the native table. The mapping may use the application name, logical table name, and a version number to create a native table key for each key in the logical table. A data structure storing these mappings may be stored at the intermediary server or in a native table in the database. This approach affords clients quick and flexible access to the database with better data integrity and security than native tables allow.