IoT Resource Knowledge Graph for Edge Computing Orchestration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resource-constrained and highly heterogeneous IoT devices face challenges in performing computing-intensive tasks due to limited computing capability and storage space, necessitating an efficient method to utilize available resources.
Innovation Solution
The method involves discovering and detecting IoT device capabilities, abstracting each device into a corresponding node based on its capabilities, and generating a resource knowledge graph to manage and orchestrate available capabilities, enabling efficient management and flexible scheduling of IoT devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If cloud computing is used to perform computing-intensive tasks, then computing capability is improved, but data transmission delay and privacy security deteriorate
Solution Approach 1:
The patent introduces an edge computing platform as an intermediary between IoT devices and cloud data centers. This platform performs computing-intensive tasks locally at the network edge, eliminating the need to transmit data to distant cloud centers while providing sufficient computing power through coordinated resource allocation across multiple IoT devices.
Solution Approach 2:
The patent transitions from a single centralized cloud computing dimension to a distributed multi-dimensional edge computing architecture. By utilizing the computing resources of multiple IoT devices across different spatial locations, the system achieves both reduced transmission delay and maintained computing capability through dimensional expansion of the computing infrastructure.
2Reliability
If edge computing is used to reduce data transmission, then privacy security is improved, but computing capability deteriorates due to resource constraints
Solution Approach 1:
The patent merges the computing resources of multiple resource-constrained IoT devices to collectively perform computing-intensive tasks. By combining CPU, GPU, and other computing resources across the network, the system achieves sufficient computing capability while keeping data processing local to maintain privacy security.
Solution Approach 2:
The patent creates a universal edge computing platform that can dynamically allocate and orchestrate various types of computing resources (CPU, GPU, storage, communication) across heterogeneous IoT devices. This multi-functional platform adapts to different task requirements while utilizing the diverse capabilities available at the network edge.
3Productivity
If heterogeneous IoT devices are utilized for edge computing, then resource utilization is improved, but system complexity deteriorates
Solution Approach 1:
The patent transforms the diverse hardware parameters of heterogeneous IoT devices into a unified resource abstraction model. By normalizing different device specifications into standardized resource units and capability descriptors, the system can efficiently allocate and manage resources across diverse devices without being overwhelmed by their heterogeneity.
Solution Approach 2:
The patent introduces a resource orchestration layer as an intermediary between the heterogeneous IoT devices and the task execution layer. This layer abstracts the complexity of device heterogeneity, providing a unified interface for resource allocation, capability matching, and task scheduling while maintaining high resource utilization across diverse devices.
Data Source
AI summary
An information processing method based on an IoT device, an information processing apparatus based on an IoT device, an information processing system based on an IoT device, an electronic device, and a storage medium are provided, where the method includes: discovering the IoT device; and detecting a capability of the IoT device; abstracting, for each IoT device, the IoT device into a corresponding node based on the capability of corresponding IoT device; generating, based on the abstracted node, a resource knowledge graph, where the node represents at least part of the capabilities of one IoT device; an edge in the resource knowledge graph represents a relationship between two adjacent nodes; and the resource knowledge graph is configured to manage and/or orchestrate the available capability on a heterogeneous IoT device.