Edge Task Scheduling via Subtask Segmentation and Dynamic Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current task scheduling methods in edge computing are inefficient due to their reliance on priority-based scheduling, leading to slow response times and low network efficiency, especially with complex and diverse user demands.
Innovation Solution
A method that splits task requirements into subtasks with constraint relationships, performs execution condition detection, and uses server state and communication information to evaluate scheduling actions through an action value evaluation model, optimizing task scheduling on edge servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If priority-based scheduling is used on a single edge server, then implementation is simple, but task execution efficiency is low and response speed is slow
Solution Approach 1:
The patent divides a complex task into multiple subtasks and schedules them on different edge servers. The task requirement is split into subtasks with constraint relationships, allowing parallel execution across multiple servers, thereby improving execution efficiency without requiring a completely complex new scheduling approach
Solution Approach 2:
The patent introduces dynamic evaluation of scheduling actions using an action value evaluation model that considers server state information, communication information, and queue information. This dynamic approach adapts to changing system conditions, improving productivity while managing complexity through intelligent decision-making
2Loss of time
If task requirements are processed as whole units, then constraint management is simple, but completion time is long
Solution Approach 1:
The patent segments task requirements into multiple subtasks that can be executed in parallel on different edge servers. By dividing the task while maintaining constraint relationships through execution condition detection, the system reduces overall completion time without losing track of task dependencies
Solution Approach 2:
The patent performs execution condition detection in advance to determine which subtasks are ready for execution. This preliminary assessment of execution conditions allows the system to proactively schedule subtasks as soon as their dependencies are met, reducing waiting time and accelerating task completion
3Productivity
If single edge server is used, then resource coordination is simple, but network efficiency is low
Solution Approach 1:
The patent combines multiple edge servers into a collaborative network for task execution. By merging the computing resources of multiple servers and coordinating them through the action value evaluation model, the system achieves higher network efficiency and resource utilization
Solution Approach 2:
The patent implements feedback mechanisms by continuously monitoring server state information, communication information, and queue information. This feedback is fed into the action value evaluation model to dynamically adjust scheduling decisions, optimizing network efficiency while managing the complexity of multi-server coordination
Data Source
AI summary
A task scheduling method includes: when a task requirement is obtained, splitting the task requirement to obtain the plurality of subtasks having a constraint relationship; performing execution condition detection on non-candidate subtasks, determining a non-candidate subtask that satisfies an execution condition as a candidate subtask, and putting the candidate subtask into a task queue; performing state detection on a server network composed of edge servers to obtain server state information and communication information; inputting the server state information, the communication information, and queue information corresponding to the task queue into an action value evaluation model to obtain the plurality of evaluated values respectively corresponding to the plurality of scheduling actions; and determining a target scheduling action from the plurality of scheduling actions by using the evaluated values, and scheduling the candidate subtask in the task queue on the basis of the target scheduling action.