Edge Computing Resource Allocation via Auxiliary Graph Pathfinding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current edge computing networks face challenges in efficiently allocating resources while ensuring minimal datacenter migration costs and meeting strict delay constraints, particularly in 5G networks that require mission-critical, delay-sensitive services for applications like IoT and auto-driving cars.
Innovation Solution
A method and system for resource allocation in edge computing networks that constructs an auxiliary graph based on predicted user routes and monitored delays, determining the lowest cost path to allocate resources across datacenters and wireless access points, ensuring assured delay constraints are met while minimizing migration costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If resources are allocated dynamically across multiple datacenters to improve service flexibility and user experience, then resource utilization and service adaptability improve, but datacenter migration costs and system complexity increase
Solution Approach 1:
The system segments the edge computing network into multiple independent datacenters, each capable of autonomous resource allocation. The resource orchestrator divides the user route into time-point segments and determines datacenter assignments for each segment independently through auxiliary graph construction, reducing overall system complexity while maintaining service flexibility.
Solution Approach 2:
The resource orchestrator acts as an intermediary between users and datacenters, managing resource allocation and migration decisions. By introducing this intermediate control layer, the system coordinates complex multi-datacenter operations without requiring direct complex interactions between all components, thereby managing system complexity while enabling flexible resource distribution.
2Loss of energy
If datacenter migration is minimized to reduce costs, then migration expenses decrease, but delay constraints may be violated and service quality deteriorates
Solution Approach 1:
The system performs preliminary actions by receiving and processing predicted user routes in advance. The auxiliary graph is constructed beforehand with all possible datacenter assignments and their associated costs and delays. This pre-computation enables the system to identify optimal migration points that satisfy delay constraints while minimizing migration costs, rather than making reactive decisions that may violate constraints or incur unnecessary costs.
Solution Approach 2:
The system incorporates feedback mechanisms by monitoring actual user routes against predicted routes and adjusting resource allocations accordingly. The resource orchestrator receives feedback on user movement and dynamically updates datacenter assignments to maintain delay constraint satisfaction while optimizing migration costs, creating a closed-loop control system that balances both objectives.
3Reliability
If strict delay constraints are enforced for mission-critical services, then service reliability improves, but resource allocation flexibility and system adaptability decrease
Solution Approach 1:
The system implements dynamic resource allocation where datacenter assignments are not fixed but adapt in real-time based on user movement and network conditions. The auxiliary graph enables dynamic re-evaluation of optimal datacenter assignments at each time point along the user route, allowing the system to maintain strict delay constraints while preserving flexibility to adapt to changing conditions through controlled re-allocation at optimal migration points.
Data Source
AI summary
Systems and methods for allocating edge computing resources may receive a predicted user route defining multiple wireless access points available along the route at respective time points and an assured delay constraint, determine, for each wireless access point, datacenters in the network meeting the assured delay constraint when accessed through the wireless access point, and construct an auxiliary graph dependent on the predicted user route and the determined datacenters. Each determined datacenter is represented by at least one node in the graph and each path in the graph defines a sequence of datacenters and a corresponding sequence of wireless access points through which they are accessible. A lowest cost path in the graph is determined and the sequences of datacenters and wireless access points defined by the lowest cost path are provided to a resource orchestrator that allocates edge computing resources for a mobile user traversing the predicted user route.