Predictive Edge Resource Allocation for Mobile Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing distance between computing devices and remote resources in distributed networks leads to higher latency in communication and processing, especially for mobile devices that move over time, causing disruptions in services like video streaming and remote processing.
Innovation Solution
A computer-implemented method and system that predict the future location of mobile devices by determining direction vectors based on historical location data, allowing for the identification and allocation of computing resources on edge nodes closer to the device's predicted location, thereby reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If remote resources are used for computing services, then processing power and storage capacity are improved, but communication latency increases due to physical distance
Solution Approach 1:
The patent segments the remote computing network into multiple distributed edge nodes geographically positioned closer to different user groups. This segmentation allows computing services to be delivered from nearby edge nodes rather than distant centralized resources, reducing communication latency while maintaining access to distributed processing power and storage capacity across the network.
Solution Approach 2:
The patent introduces edge nodes as intermediary computing resources positioned between end-user devices and centralized cloud resources. These intermediaries cache and process data locally at edge locations, reducing the need for long-distance communications while still leveraging remote resources when necessary, thus balancing processing power access with latency reduction.
2Loss of time
If edge nodes are deployed closer to users, then communication latency is reduced, but network complexity and infrastructure cost increase
Solution Approach 1:
The patent designs edge nodes with multi-functional capabilities that can perform diverse computing tasks including caching, data processing, analytics, and service delivery. This universality allows a single edge node infrastructure to handle multiple functions, reducing the need for specialized components and simplifying network complexity while maintaining low-latency performance across different service types.
Solution Approach 2:
The patent implements predictive resource allocation at edge nodes using machine learning models that anticipate user requests and pre-position data and computing resources before they are needed. This preliminary action reduces actual service delivery latency while optimizing edge node utilization, preventing over-provisioning and reducing network complexity through intelligent resource management.
3Productivity
If computing resources are allocated dynamically based on demand, then resource utilization efficiency is improved, but response time and control complexity increase
Solution Approach 1:
The patent employs machine learning models to predict future computing service demands and proactively allocates resources at edge nodes before requests arrive. This preliminary resource allocation based on predictions maintains high resource utilization efficiency while avoiding the latency associated with dynamic resource provisioning at request time, as resources are already prepared and positioned.
Solution Approach 2:
The patent implements feedback mechanisms where edge nodes continuously monitor actual service delivery performance and resource utilization, feeding this information back to the resource allocation system. This feedback loop enables the system to refine its predictions and adjust resource allocation strategies, maintaining high efficiency while keeping response times low through data-driven optimization.
Data Source
AI summary
The present technology relates to improving computing services in a distributed network of remote computing resources, such as edge nodes in an edge compute network. In an aspect, the technology relates to a method that includes aggregating historical request data for a plurality of requests, wherein the aggregated historical request data a time of the request, a location of a device from which the request originated, and/or a type of service being requested. The method also incudes training a machine learning model based on the aggregated historical request data; generating, from the trained machine learning model, a prediction for a type of service to be request; identifying an edge node, from a plurality of edge nodes, based on a physical location of the edge node; and based on predicted service, allocating computing resources for the computing service on the identified edge node.