Adaptive IoT Application Unloading in Edge Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for computing unloading in IoT devices face challenges in decoupling strongly coupled applications into functional modules and optimizing resource allocation across IoT equipment, edge servers, and cloud servers, leading to inefficiencies and increased overhead.
Innovation Solution
A method is introduced that uses a universal program structure supporting on-demand computing unloading, combined with a multi-task particle swarm optimization-genetic algorithm (MPSO-GA) to analyze the MEC environment and determine optimal unloading solutions, minimizing system overhead while satisfying deadline constraints for each application.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If computing tasks are unloaded to cloud server, then computing capacity is improved, but execution delay increases and network traffic congestion occurs
Solution Approach 1:
The patent segments the monolithic cloud computing system into a hierarchical structure with edge servers positioned between IoT devices and cloud servers. Computing tasks are divided and distributed to appropriate layers: time-sensitive tasks are offloaded to edge servers for rapid processing, while non-time-critical tasks can be handled by cloud servers, thus reducing execution delay while maintaining computing capacity.
Solution Approach 2:
Edge servers act as intermediaries between IoT devices and cloud servers. They provide local computing resources that reduce the distance for data transmission, thereby decreasing execution delay and network traffic congestion while still enabling access to cloud computing capacity when needed.
2Loss of time
If computing tasks are unloaded to edge server, then execution delay is reduced, but system complexity increases
Solution Approach 1:
The patent designs a universal program structure that can operate in multiple modes: local execution on IoT devices, offloading to edge servers, or further offloading to cloud servers. This multi-functional architecture allows the same application framework to adapt to different deployment scenarios without requiring separate systems, thereby managing complexity while providing flexible delay optimization.
Solution Approach 2:
The system employs dynamic unloading decision-making that adapts to varying conditions. The unloading strategy is not fixed but dynamically adjusts based on task characteristics, edge server availability, and network conditions, allowing the system to optimize execution delay without requiring complex static configurations for every possible scenario.
3Adaptability or versatility
If application is divided into fine granularity, then unloading flexibility is improved, but overhead increases
Solution Approach 1:
The patent implements service-level granularity for unloading decisions rather than requiring fine-grained decomposition of all application components. This partial action approach unloads only the necessary services to edge or cloud servers while keeping other components local, achieving adequate flexibility without the excessive overhead that would result from unloading every individual function.
4Productivity
If multiple applications share edge resources, then resource utilization is improved, but unloading decision difficulty increases
Solution Approach 1:
The patent changes the decision parameters from considering individual application characteristics to using service-level metrics and aggregate resource utilization. The unloading decision-making algorithm operates at the service level with simplified parameters, allowing multiple applications to share edge resources efficiently without requiring complex analysis of each application's internal structure, thus improving resource utilization while managing decision difficulty.
Data Source
AI summary
A method for supporting adaptive unloading of multi-Internet of Things (IoT) applications in an edge environment includes: constructing an application compliant with a universal program structure supporting on-demand computing unloading; for the application compliant with the universal program structure supporting on-demand computing unloading, extracting a program fragment flowchart through static code analysis to provide internal flow information of the application for generation of an unloading solution; analyzing a peripheral mobile edge computing (MEC) environment and the application through an unloading solution generating algorithm based on a multi-task particle swarm optimization-genetic algorithm to obtain the optimal unloading solution; and performing, by the application, computing unloading by taking a service as a particle size according to the unloading solution to minimize a system overhead under a circumstance that a deadline constraint of each application is satisfied. The method supports the computing unloading of different types of applications in the MEC environment.