Distributed Computing Management for Edge Task Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network architectures struggle to provide agile computing and communication mechanisms for applications and services like 5G and 6G, leading to suboptimal performance due to separate communication and computing networks.
Innovation Solution
Integrating communication and computing resources into a unified network fabric, utilizing a distributed intelligent layer to identify and offload computing tasks to nodes within the network edge without traversing the core network, employing techniques for registering, discovering, and matching computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data traverses through the core network for computing tasks, then network coverage and connectivity are ensured, but network latency increases and performance deteriorates
Solution Approach 1:
The patent segments the network architecture into core network functions and edge computing functions. The DCMF is separated from the core network and deployed at the network edge, allowing computing tasks to be processed locally without traversing the entire core network path. This segmentation reduces transmission latency while maintaining network management capabilities.
Solution Approach 2:
The patent introduces a new dimensional layer - the distributed computing management function layer - that operates between the user equipment and the core network. This intermediate layer enables direct peer-to-peer computing task offloading and data processing at the edge, creating a parallel processing dimension that bypasses the traditional sequential core network path.
2Productivity
If computing resources are centralized in the core network, then resource management is simplified, but computing agility and task offloading efficiency decrease
Solution Approach 1:
The DCMF implements self-service capabilities by autonomously discovering available computing resources at the network edge, automatically matching computing tasks with suitable nodes, and dynamically managing resource allocation. This eliminates the need for manual resource provisioning while maintaining efficient resource utilization through automated decision-making algorithms.
Solution Approach 2:
The system establishes feedback loops where the DCMF continuously monitors computing resource status, task performance metrics, and network conditions. Based on this feedback, the DCMF dynamically adjusts resource allocation, task routing, and node selection to optimize computing efficiency while simplifying management through adaptive control.
3Adaptability or versatility
If a unified network fabric integrating communication and computing is deployed, then computing agility improves, but network architecture complexity increases
Solution Approach 1:
The DCMF serves multiple functions within a single architectural component: it manages computing resource discovery, task offloading coordination, node matching, and resource allocation. This multi-functionality integrates communication and computing management capabilities into a universal platform, improving agility without proportionally increasing complexity.
Solution Approach 2:
The DCMF acts as an intermediary layer between the communication network and computing resources. It translates communication requirements into computing task specifications and coordinates resource allocation without requiring direct integration of all network and computing components. This intermediary approach enables unified network fabric functionality while managing complexity through abstraction.
Data Source
AI summary
A distributed computing management function (DCMF) is configured to receive computing resource availability information from a first network node and transmit computing node information associated with the first network node to a second network node, wherein the second network node offloads one or more computing tasks to the first network node and the data for the one or more computing tasks does not traverse a core network.