Dynamic Cloud BPM Task Assignment via Competitive Bidding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional business process management (BPM) systems struggle in cloud computing environments due to static workflow configurations, inefficient data transfer, and lack of optimization across multiple service providers, leading to increased overhead and potential performance breakdowns.
Innovation Solution
A dynamically optimized distributed cloud computing-based BPM system that utilizes a cloud registry server and cloud workflow manager to dynamically assign tasks to the most efficient service providers, allowing for competitive bidding and real-time optimization of workflow execution based on current service provider conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If a conventional BPM system uses fixed computing components/resources from internal enterprise networks, then system stability is maintained, but adaptability to cloud computing environments deteriorates
Solution Approach 1:
The workflow engine is designed to dynamically select and switch between internal enterprise computing resources and external cloud computing resources based on real-time conditions. The system maintains a pool of both internal and external service providers, allowing the workflow to be executed on whichever resources are most appropriate at any given moment, thus achieving both stability and adaptability
Solution Approach 2:
The BPM system is designed to universally operate with multiple types of computing resources - both internal enterprise resources and external cloud resources - through a unified interface. The workflow engine can treat different resource types uniformly while maintaining their individual characteristics, enabling the system to function effectively in hybrid cloud environments
2Device complexity
If the workflow engine acts as a centralized switchboard operator managing all data transfers, then coordination is simplified, but data transfer overhead increases
Solution Approach 1:
The system segments the workflow execution into independent tasks that can be distributed to different service providers. Each task can be executed autonomously by the assigned service provider without requiring constant coordination through the workflow engine, reducing the engine's coordination burden and minimizing data transfer overhead through direct peer-to-peer communication channels
Solution Approach 2:
The workflow engine acts as an intermediary that initially assigns tasks to service providers but then allows direct communication between service providers for subsequent data exchanges. This mediator approach reduces overhead by enabling service providers to communicate directly rather than routing all communications through the engine
3Ease of manufacture
If static conditions are used in workflow construction with explicit service providers, then workflow definition is simplified, but optimization across multiple service providers deteriorates
Solution Approach 1:
The system dynamically selects service providers for workflow tasks based on real-time conditions such as service provider availability, performance metrics, and cost considerations. The workflow engine maintains a pool of service providers and automatically selects the most appropriate one for each task, enabling optimization without complicating workflow definition
Solution Approach 2:
Service providers autonomously respond to workflow task requests and self-select for tasks based on their capabilities and current state. The system allows service providers to independently manage their own workloads and capabilities, enabling automatic optimization without requiring complex centralized assignment logic
4Productivity
If multiple workflow engines are used to distribute workflows, then load balancing is improved, but coordination complexity and reliability deteriorate
Solution Approach 1:
The system segments workflow execution into independent, self-contained tasks that can be distributed to different service providers. Each task is sufficiently independent that it can be executed without requiring complex coordination between multiple workflow engines, allowing load balancing while minimizing coordination overhead
Solution Approach 2:
The system implements feedback mechanisms where service providers report their status, performance, and availability to the workflow engine. This feedback enables the engine to dynamically balance load across available resources and adjust task assignment in real-time, achieving load balancing through simple reactive control rather than complex proactive coordination
Data Source
AI summary
A workflow server can receive requests, each for a business process workflow conforming to a business process model. Each business process workflow can include a set of interdependent tasks. The workflow server can satisfy received requests by assigning tasks to different service providers that provide software services. Each of the tasks can be assigned to corresponding ones of the software services. For each task, the workflow server can also defines an allocated cost per software service, and a time allocation per software service for completing the corresponding one of the tasks. Different service providers, including those assigned to tasks, can receive information for ones of the tasks not directly assigned to them by the workflow server. The different service providers can then bid on these tasks. Wherein when bids are won, tasks for a business process flow can be reassigned based on winning bids.