Edge Cloud Container Provisioning from Network Event Hints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resource allocation in edge cloud platforms is challenging due to their small size and the variety of applications they need to support, leading to inefficiencies in managing application containers.
Innovation Solution
A network API service that receives network event data from access and core networks to provide performance hints for efficient management of edge cloud resources, including booting, migrating, or shutting down application containers based on predicted user behavior and network events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If edge cloud locations maintain small size to reduce deployment cost and complexity, then device complexity and deployment cost are reduced, but resource allocation efficiency deteriorates due to limited resources and variety of applications
Solution Approach 1:
The system performs preliminary actions by predicting user equipment arrival at edge cloud locations before actual occurrence. Network event data is analyzed in advance to generate performance hints about upcoming user arrivals, allowing the edge cloud resource manager to pre-provision application containers and allocate resources proactively, thus improving resource allocation efficiency while maintaining small edge cloud sizes
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring network event data from access and core networks. This feedback loop provides real-time information about user equipment behavior and network conditions, enabling dynamic adjustment of resource allocation decisions at edge cloud locations, thereby optimizing resource efficiency without increasing complexity
2Adaptability or versatility
If application containers are kept ready to support variety of applications, then adaptability is improved, but resource waste increases due to idle containers consuming limited edge cloud resources
Solution Approach 1:
The system applies preliminary action by predicting which application containers will be needed based on analyzed network event data. Instead of keeping all possible containers ready, the system provisions only those containers that are predicted to be required soon, thereby maintaining adaptability for supporting various applications while avoiding the resource waste of keeping idle containers running
Solution Approach 2:
The system changes the operational state parameters of application containers dynamically based on predicted user behavior. Containers are instantiated, migrated, or terminated based on performance hints derived from network event data, allowing the system to adapt application availability to actual demand patterns and reduce resource consumption from idle containers
3Loss of time
If application containers are instantiated quickly to improve user-perceived response times, then response time is reduced, but resource management complexity increases
Solution Approach 1:
The system reduces user-perceived response time by performing preliminary provisioning actions. Application containers are predicted to be needed and are instantiated or migrated to edge cloud locations in advance of actual user requests, based on analysis of network event data. This proactive approach ensures containers are ready when needed, minimizing latency without requiring complex on-demand instantiation mechanisms
Data Source
AI summary
Aspects of the subject disclosure may include, for example, a network API service that receives network event data and provides performance hints to a resource manager that manages application containers at edge cloud locations. Network event data may be received from access networks, core networks, nodes within access networks or core networks, or the like. Performance hints may allow booting of application containers at edge cloud locations. Other embodiments are disclosed.