State-Aware Data Model for Multi-State Persistence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data modeling techniques require storing multiple copies of data models for each state in a data storage system, leading to redundant data storage, high performance costs due to repeated change propagation, and limitations in multi-tenancy within the persistence layer.
Innovation Solution
A state-aware data model is generated that represents data sets as entities, relationships, and properties, with a state-aware attribute indicating which states each entity and association belong to, allowing a single copy of an entity or relationship to be associated with multiple states and propagating changes efficiently across states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If multiple full copies of data models are stored for each state, then rapid access to data model for a given state is achieved, but a large amount of duplicate data is stored
Solution Approach 1:
The patent merges multiple state representations into a single shared data model structure. Instead of storing separate full copies for each state, the invention uses a unified model where states are represented through version identifiers and change sets, allowing multiple states to coexist in one consolidated structure that reduces duplicate data storage.
Solution Approach 2:
The patent creates a universal data model structure that serves multiple states simultaneously. The single data model can represent multiple states through versioning mechanisms, where one structure performs the function of storing and managing multiple state representations, eliminating the need for separate dedicated databases for each state.
2Reliability
If changes are propagated forward through multiple states, then state consistency is maintained, but a huge performance cost is incurred due to repeated application
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing change sets between states. Instead of propagating changes sequentially through each state when needed, the invention pre-computes the differences between states and stores them for direct application, eliminating repeated propagation operations and their associated performance costs.
Solution Approach 2:
The patent uses copying by creating lightweight change set representations rather than repeatedly applying full change propagation sequences. The invention copies only the necessary differential information between states, allowing rapid state transitions without the computational overhead of repeated full propagation operations.
3Speed
If each state is persisted to a dedicated database, then rapid access to state data is enabled, but multi-tenancy within the persistence layer is obstructed
Solution Approach 1:
The patent merges multiple state persistences into a single shared database structure. Instead of dedicating separate databases to each state, the invention uses one persistent storage system with versioning and state identification mechanisms that allow multiple states to be stored and accessed efficiently, thereby enabling multi-tenancy while maintaining access performance.
Data Source
AI summary
A plurality of data sets is identified. A data model is generated representing the data sets. The data model represents the data sets and variations of the data sets as entities, relationships between the data sets as associations, and values associated with the entities as properties. The data model identifies states corresponding to the data sets wherein each state has associated therewith one or more of the entities, the associations and the properties. A state-aware attribute is maintained for each entity and for each association indicating to which one or more of the states the entity and the association respectively belong.