Global Resource Scheduling Using Provider Status Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network scheduling strategies lack global optimization, leading to inefficient routing of service requests and suboptimal utilization of computing resources due to lack of awareness between network and data center status.
Innovation Solution
A resource scheduling method that considers a global optimization goal to select the optimal target resource provider based on service requirements and provider status, using incentives to align with global optimization objectives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If single-time user service request scheduling is used, then user service request optimization is achieved, but global performance optimization is not realized
Solution Approach 1:
The patent introduces a feedback mechanism where resource providers report their status and performance data back to the scheduling system. This feedback loop enables the system to make globally optimized scheduling decisions based on real-time information about resource availability, service quality, and network conditions, resolving the contradiction between single-time optimization and global optimization.
Solution Approach 2:
The patent transitions from single-time, single-user scheduling to multi-dimensional scheduling that considers global optimization goals across multiple dimensions including time, resources, network conditions, and service quality. This dimensional expansion enables simultaneous optimization of user requests and global performance.
2Productivity
If network and data center status are not shared, then system simplicity is maintained, but service request routing efficiency decreases
Solution Approach 1:
The patent introduces an intermediary scheduling system that acts as a mediator between the network and data center. This intermediary collects status information from both sides, processes it according to global optimization goals, and makes informed routing decisions, thereby improving efficiency without requiring direct complex integration between network and data center.
Solution Approach 2:
The system performs preliminary actions by pre-collecting and processing status information from network and data center components. This preliminary data gathering and processing enables faster, more efficient routing decisions when service requests arrive, improving productivity without adding real-time processing complexity.
3Productivity
If global optimization goals are implemented, then resource service efficiency is improved, but incentive mechanism complexity increases
Solution Approach 1:
The patent uses parameter changes by adjusting incentive parameters based on global optimization goals. The system dynamically modifies incentive values and weighting factors to balance different optimization objectives such as resource utilization, service quality, and cost efficiency, thereby achieving improved productivity through manageable parameter adjustments rather than complex mechanisms.
Data Source
AI summary
The present disclosure provides a resource scheduling method, device and apparatus. The method includes: receiving a service request, where the service request carries service requirement; generating a resource provider list of resource providers that meet the service requirement based on the service requirement and status of the resource providers; selecting a target resource provider in the resource provider list based on the resource provider list and first information; wherein the first information is used to indicate a global optimization goal; and calling the target resource provider to provide service.