Distributed Task Scheduling Across Computing Nodes for Resource Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computing power networks, network computing power utilization and transmission resource utilization are low during the computing task scheduling process.
Innovation Solution
A computing task scheduling method that divides tasks into subtasks, determines computing and communication requirements, and schedules them to appropriate computing nodes, improving resource utilization and ensuring service experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computing tasks are scheduled in a computing power network, then service functionality is provided, but network computing power utilization and transmission resource utilization remain low
Solution Approach 1:
The computing task is divided into multiple subtasks that can be distributed to different computing nodes for parallel execution. This segmentation enables better utilization of network computing power resources while maintaining service functionality through coordinated subtask execution.
Solution Approach 2:
The patent introduces a new dimension of task decomposition and distribution across multiple computing nodes, transforming the scheduling approach from monolithic task execution to distributed subtask coordination, thereby improving resource utilization without compromising service quality.
2Productivity
If computing tasks are divided and scheduled to multiple computing nodes, then resource utilization improves, but scheduling complexity increases
Solution Approach 1:
The computing management function acts as an intermediary that receives computing task requirements from the first apparatus, divides tasks into subtasks, and schedules them to appropriate computing nodes. This intermediary approach simplifies the overall scheduling complexity by centralizing the decision-making process while enabling distributed execution.
Solution Approach 2:
The computing management function performs preliminary task division and scheduling decisions before actual execution, determining the allocation of subtasks to computing nodes in advance. This preliminary action reduces runtime complexity and enables more efficient resource utilization.
Data Source
Figure 1
Figure 2(a)~2(b)
Figure 3
AI summary
A computing task scheduling method and a communication apparatus are provided, to improve network computing power utilization and transmission resource utilization in a computing task scheduling process. The method includes: receiving first information, where the first information indicates a requirement on a computing task of a first apparatus, and the computing task includes at least one subtask; and sending second information in response to the first information, where the second information indicates one or more subtasks in the at least one subtask scheduled to a first computing node.