Dynamic Pricing Model for Edge Computing Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The growth of IoT devices with demanding latency requirements and security concerns in telecommunications networks necessitates an optimized method for resource allocation and pricing, as edge computing resources are distributed and may be in short supply at specific locations, requiring sophisticated management to meet diverse service demands.
Innovation Solution
A dynamic, real-time pricing model using a double auction mechanism for automatic resource management, which considers current demand, resource availability, and historical usage to efficiently allocate and charge for edge computing resources, ensuring revenue maximization and cost minimization for network operators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If edge computing resources are distributed throughout the network, then latency requirements are met and security is improved, but resources may be in short supply at specific locations requiring sophisticated management
Solution Approach 1:
The patent implements automated resource management where the system autonomously performs demand function generation, supply function generation, and allocation determination without manual intervention. The double auction model automatically matches customer demands with available resources, and the service conductor self-adjusts allocations based on changing conditions, eliminating the need for manual resource vetting and evaluation.
Solution Approach 2:
The system dynamically adjusts allocation parameters and pricing based on real-time conditions. The demand function and supply function are continuously updated with current network resource usage and availability data, allowing the allocation determination to adapt to changing demands and resource states, thereby optimizing resource distribution across the network.
2Ease of operation
If manual resource management is used, then resource allocation can be controlled, but it requires manual vetting and evaluation of different use cases which is inefficient
Solution Approach 1:
The patent replaces manual mechanical resource management processes with an automated computational system. The service conductor uses algorithmic demand functions, supply functions, and double auction models to automatically determine resource allocations, substituting human manual vetting and evaluation with automated mathematical optimization processes that are both efficient and highly automated.
3Quantity of substance
If resources are concentrated in a central repository, then resource availability is improved, but latency requirements cannot be met for time-sensitive applications
Solution Approach 1:
The patent segments the centralized resource pool into distributed edge computing resources located at multiple network nodes. The service conductor manages these segmented resources locally at each edge location, allowing resources to be physically closer to data sources and consumers, thereby reducing transmission latency while maintaining overall resource availability through the distributed architecture.
Solution Approach 2:
The system transitions from a single-dimensional centralized resource model to a multi-dimensional distributed edge model. Resources are allocated across multiple spatial dimensions (different edge locations) and temporal dimensions (dynamic allocation over time), enabling simultaneous optimization of both resource availability and latency by serving different demands from optimally located edge resources.
Data Source
AI summary
A system, method and computer readable storage medium are disclosed to provide dynamic network resource management in a telecommunications network where at least a portion of the network resources are located at an edge of the telecommunications network. The capacity available, the aggregate demand of resources at a given time, and the cost of the network resource demanded as well as each customers upper price limit may be considered and adjusted in real-time by different network resource allocation models.