Edge Computing Load Sharing for Response Time Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional workload distribution models in cloud and edge computing lead to data transfer delays and inefficient resource utilization due to inadequate communication capacity and improper workload processing between cloud and edge devices, resulting in suboptimal response times for users.
Innovation Solution
A cloud load sharing system that predicts edge device locations and communication capacities, shifts cloud workloads to edge devices during processing, and installs necessary logical functions to minimize data transfer delays and optimize response times by leveraging edge device resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If cloud applications process workloads remotely through conventional cloud computing, then centralized processing capability is improved, but data transfer delay increases due to communication capacity limitations
Solution Approach 1:
The patent segments the cloud application workload into multiple logical functions that can be distributed and executed across different locations. Instead of requiring all processing to occur centrally in the cloud, the system divides the workload so that edge devices can execute portions of the application locally, reducing the amount of data that needs to be transferred back and forth with the cloud server.
Solution Approach 2:
The patent implements preliminary installation of cloud applications on edge devices before actual workload execution. The system predicts future workloads and pre-installs necessary logical functions on edge devices in advance, enabling them to process workloads locally without requiring real-time data transfer from the cloud, thus reducing latency.
2Speed
If edge devices process real-time data locally, then response speed is improved, but resource utilization efficiency deteriorates due to inconsistent workload distribution
Solution Approach 1:
The patent implements dynamic workload distribution where the system continuously monitors edge device resources and communication conditions. The workload distribution model adapts in real-time, shifting between cloud-processing and edge-processing modes based on current system state, ensuring optimal resource utilization while maintaining fast response times.
Solution Approach 2:
The system changes the parameter of workload execution location based on predicted communication capacity and device resources. When communication conditions are poor or devices have sufficient capacity, the system shifts workloads from cloud to edge execution, optimizing resource utilization while maintaining response performance.
3Device complexity
If conventional workload distribution is used with fixed cloud-edge separation, then system simplicity is maintained, but adaptability to varying communication conditions deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors actual communication capacity and device resource usage. Based on this feedback, the system adjusts the workload distribution strategy in real-time, transitioning between fixed cloud-edge separation and dynamic load sharing modes to adapt to varying communication conditions.
Solution Approach 2:
The system performs preliminary analysis of predicted communication capacity and device resources before workload execution. This advance preparation enables the system to pre-determine optimal workload distribution strategies, combining the simplicity of fixed models with the adaptability of dynamic adjustment by predicting future conditions in advance.
Data Source
AI summary
Methods, computer program products, and systems are presented. The methods include, for instance: analyzing resources of an edge device available for computing workloads of a cloud to which the edge device is operatively coupled various communication networks per locations of the edge device, wherein the edge device is mobile. A location of the edge device at an estimated time of delivery of an output of a cloud application is predicted prior to the estimated time of delivery. It is determined that the location of the edge device from the predicting is serviced by a communication network below a threshold connectivity. The cloud has the cloud application installed on the edge device at a current location according to an access permission on the edge device. The edge device continues processing the workloads of the cloud and the output of the cloud application generated.