Multi-Cloud Task Allocation Using Shortest-Path Broker Pricing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Establishing a multi-cloud environment is computationally complex and inefficient, particularly for large markets, leading to economic losses for both cloud providers and buyers, and there is little incentive for parties to disclose confidential information.
Innovation Solution
A computer-implemented method involving a central broker that constructs a graph of nodes and edges based on cost parameters to determine the shortest path distance for task allocation among participants, using Integer Linear Programming to optimize task allocation and payment, ensuring truthful declarations and minimizing total cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a multi-cloud environment is established using conventional methods, then cloud services can be provided across multiple platforms, but the computational complexity increases significantly and economic losses occur
Solution Approach 1:
A central broker is introduced as an intermediary entity that receives cost parameter declarations from cloud providers and buyers, constructs graphs to model the multi-cloud environment, and determines optimal task allocations and payments. This mediator simplifies the complex interactions between multiple cloud platforms by centralizing the computation and coordination functions.
Solution Approach 2:
The multi-cloud environment is segmented into distinct components: cloud providers declaring cost parameters, cloud buyers requesting tasks, a central broker managing the marketplace, and task allocation mechanisms. This segmentation allows each component to operate independently while contributing to the overall system, reducing the computational burden on any single entity.
2Productivity
If conventional methods are used to establish multi-cloud environments, then task allocation can be performed, but the process is inefficient and time-consuming
Solution Approach 1:
Cloud providers预先 declare their cost parameters and capabilities to the central broker before tasks are assigned. This preliminary action allows the broker to pre-construct graphs and pre-calculate optimal allocations, significantly reducing the time required when actual task allocation needs to occur.
Solution Approach 2:
The patent replaces complex iterative optimization algorithms with graph-theoretic approaches. By modeling the multi-cloud environment as graphs with nodes representing cloud platforms and edges representing connections, the system can efficiently determine optimal task allocations using shortest path algorithms, substituting computationally intensive mechanical optimization with more elegant mathematical modeling.
3Measurement precision
If parties disclose confidential cost information, then optimal task allocation can be achieved, but there is little incentive for participants to be transparent
Solution Approach 1:
The system employs self-service mechanisms where cloud providers and buyers independently declare their own cost parameters and capabilities without external verification or pressure. The central broker accepts these declarations and uses them to construct graphs and determine allocations, allowing participants to maintain control over their confidential information while still enabling optimal system-wide optimization.
Data Source
AI summary
Described are techniques for establishing a multi-cloud environment. A declared type ci is received from a participant i (e.g., cloud buyer) corresponding to cost parameters. The allocation of tasks among the participants (e.g., cloud buyers, cloud providers) is obtained by solving an optimization problem. A graph of nodes and edges is then constructed, where the nodes include a super node and standard nodes corresponding to the possible types of the participant i. Payment from the participant i to the central broker involving the allocation of tasks among the participants is determined corresponding to the shortest path distance from the super node to ci. A multi-cloud environment with cloud providers and cloud buyers is then created based on the allocation of the tasks among the participants and based on the payment between the central broker and the cloud buyers/cloud providers.