Computing Force Network Path Scheduling With Node Influence Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for scheduling computing force network resources fail to explicitly select resources based on task requirements, neglect transport nodes, and lack comprehensive evaluation of end-to-end links, leading to suboptimal path selection and increased complexity.
Innovation Solution
A method for resource scheduling in a computing force network that evaluates node and forward path performance by considering resource indicators and influence nodes, using a comprehensive evaluation function to determine optimal paths that meet specific business requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing scheduling solutions are used, then resource allocation is performed, but path selection is suboptimal and complexity increases
Solution Approach 1:
The patent segments the resource scheduling process into distinct modules: node performance evaluation module, forward path performance evaluation module, and optimal path selection module. Each module handles specific evaluation tasks independently, improving path selection quality while managing complexity through modular design. The node evaluation module assesses individual node metrics, the forward path module evaluates complete paths, and the selection module chooses optimal paths based on business requirements.
Solution Approach 2:
The patent introduces comprehensive evaluation functions that transform multiple resource indicators (computing power, storage, bandwidth, latency) into standardized performance indicators. This parameter transformation enables quantitative comparison of different paths by converting heterogeneous resource metrics into unified evaluation dimensions, thereby improving path selection quality while maintaining manageable system complexity through standardized assessment criteria.
2Measurement precision
If comprehensive node and path evaluation is implemented, then optimal path selection improves, but computational complexity increases
Solution Approach 1:
The evaluation process is divided into two sequential stages: first, node performance evaluation where individual node metrics are assessed and standardized; second, forward path performance evaluation where complete paths are evaluated based on composed node indicators. This segmentation allows precise measurement at each level while managing overall complexity through hierarchical processing, avoiding the need to evaluate all possible paths simultaneously.
Solution Approach 2:
The patent performs preliminary node performance evaluation before conducting forward path evaluation. By pre-computing and standardizing node performance indicators first, the system prepares evaluation data in advance, enabling more accurate path assessments without repeating node-level computations for each path. This preliminary action reduces redundant calculations and manages evaluation complexity while maintaining high measurement precision.
3Manufacturing precision
If influence nodes are considered in evaluation, then resource allocation accuracy improves, but system complexity increases
Solution Approach 1:
The patent applies local quality evaluation by considering influence nodes specific to each evaluated node rather than uniformly evaluating all nodes with the same criteria. Each node's performance indicator incorporates characteristics of its relevant influence nodes (such as upstream or downstream nodes in the forward path), enabling accurate resource allocation decisions tailored to local network conditions while managing overall system complexity through targeted, context-specific evaluations.
Data Source
AI summary
A method of resource scheduling in a computing force network, wherein the computing force network includes at least one node among which at least one forward path exists, and the method includes the following steps: for each of the at least one node, acquiring node performance indicator of the node based on resource indicators of the node and at least one influence node related to the node; for each of the at least one forward path, acquiring path performance indicator of the forward path based on the performance indicators of nodes constituting the forward path; and in response to a business requirement for the computing force network, selecting at least one optimal forward path from the at least one forward path based on the forward path performance indicators of the at least one forward path.


