Edge Application Partitioning for Hardware-Aware Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing edge computing systems lack efficient methods for optimizing the deployment and resource allocation of applications across edge devices, leading to suboptimal performance and resource utilization.
Innovation Solution
A computer-implemented edge utility management method that receives applications for deployment on edge devices, obtains application and hardware profiles, and deploys application parts to selected hardware resources based on suitability indicators, facilitating holistic access to hardware resources and communication between deployed parts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If applications are deployed on edge devices without optimized resource allocation, then deployment simplicity is maintained, but resource utilization efficiency deteriorates
Solution Approach 1:
The system performs preliminary actions by obtaining application profiles and hardware profiles before deployment, analyzing execution requirements and resource capabilities in advance, and creating a suitability mapping that guides optimal resource allocation during deployment
Solution Approach 2:
The patent introduces an intermediary management system that acts as a mediator between applications and hardware resources. This intermediary obtains profiles, analyzes compatibility, determines suitability indicators, and facilitates optimal matching without requiring direct complex interactions between applications and hardware
2Productivity
If hardware resources are accessed without holistic access mechanisms, then system simplicity is maintained, but resource allocation optimality deteriorates
Solution Approach 1:
The patent implements a universal access mechanism that can access hardware resources across different operating systems and hardware types through a single interface. The hardware profile abstraction layer provides multi-functionality, enabling the same access mechanism to work with diverse hardware resources including processors, storage, and specialized accelerators
3Manufacturing precision
If application parts are not analyzed for execution requirements, then deployment speed is maintained, but deployment accuracy deteriorates
Solution Approach 1:
The system performs preliminary analysis of application parts to obtain execution requirements before deployment. By analyzing metadata, tags, and execution preferences in advance and storing them in application profiles, the system prepares deployment information beforehand, enabling quick and accurate deployment decisions without time-consuming analysis during the actual deployment process
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a computer-implemented edge utility management method, the method comprising: receiving an application (200) to be deployed on an edge device (100), wherein the application comprises a plurality of separately-deployable application parts (202-206); obtaining application profiles for one or more of the applications parts; obtaining hardware profiles for one or more hardware resources of the edge device; deploying the application parts (202-206) to selected hardware resources based on the application profiles and the hardware profiles; and facilitating communication between the deployed application parts during execution of the application.