Persistenceless BPM System State Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Business Process Management (BPM) systems face issues with data accuracy and system performance due to the constant synchronization of state and business context data across multiple data stores, leading to unreliable data and increased complexity in system recovery upon failures.
Innovation Solution
A persistenceless BPM system that delegates state data management to client applications, using event message flows to communicate process state transactions, allowing state data to be stored alongside business context data and enabling efficient restoration of business process instances from external data stores in case of system failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional BPM systems store state data at the BPM system and business context data at client-side applications, then the system can track process states in real time, but the BPM system and client applications must constantly synchronize and merge information across multiple data stores, consuming processing power and network bandwidth, and increasing system complexity
Solution Approach 1:
The patent merges state data and business context data into a single data store located at the client application. This eliminates the need for constant synchronization between multiple data stores, reducing system complexity while maintaining data accuracy. The BPM system can still track process states by receiving notifications from the client application when state changes occur.
Solution Approach 2:
The patent introduces a notification mechanism as an intermediary between the client application and BPM system. When state data changes at the client application, a notification is sent to the BPM system, which then updates its tracking without requiring full synchronization. This mediator approach reduces the synchronization overhead while maintaining real-time tracking capability.
2Reliability
If conventional BPM systems constantly synchronize state data and business context data across multiple data stores, then the system can maintain real-time tracking, but processing power and network bandwidth are heavily consumed by synchronization operations
Solution Approach 1:
Instead of continuous synchronization, the patent implements periodic action by only triggering data synchronization when state changes occur. The client application sends notifications to the BPM system at these discrete change points, eliminating continuous processing overhead while maintaining accurate real-time tracking of process states.
Solution Approach 2:
The client application takes responsibility for maintaining its own state data locally and only communicates changes to the BPM system. This self-service approach eliminates the need for the BPM system to continuously query and synchronize data, significantly reducing processing power consumption on both ends while maintaining data accuracy.
3Adaptability or versatility
If conventional BPM systems use multiple physically and/or logically separated data stores, then the system can separate state management and business context management, but the complexity of system recovery upon failure increases and recovery procedures may become inaccurate or fail
Solution Approach 1:
The patent combines state data and business context data into a single data store at the client application. This merger simplifies system recovery because there is only one data store to restore rather than multiple separated stores that must be synchronized. The BPM system can recover by receiving a single notification from the client application, eliminating the complexity of coordinating recovery across multiple data stores.
4Reliability
If conventional BPM systems store all state data at the BPM system, then the system can centrally manage process state, but the BPM system consumes more processing power and network bandwidth for synchronization operations
Solution Approach 1:
The patent extracts the state data storage function from the BPM system and places it at the client application. The BPM system retains only the process state tracking function, receiving notifications from the client application when state changes occur. This extraction reduces the processing power and network bandwidth requirements of the BPM system while maintaining reliable state tracking through the notification mechanism.
Data Source
AI summary
Embodiments of the present invention allow for a persistenceless BPM system capable of operating with external state and business context data stores managed by client-side applications. Operating without a state data store within the system, the persistenceless BPM system delegates the responsibility of managing and storing of the state persistence data to the client applications and relies on event message flows to communicate all relevant process state transactions to the client applications. Accordingly, state data is stored in the same data store as the relevant business context data. In the event of a system failure, the process instances are restored from the outsourced data store managed by the client application. Recoverability of operations is improved as there is no need for data reconciliation between two or more data distinct stores. Overall, embodiments of the present invention improve system performance and scalability as many unnecessary costly operations are eliminated either partially or completely.