Edge Resource Reservation for Reliable EAS Deployment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The deployment of edge application servers (EAS) in edge data networks (EDN) often fails due to insufficient remaining resources, leading to high deployment failure rates and increased latency when the 'first come, first served' mechanism is used.
Innovation Solution
A method for reserving resources in advance by an edge computing service provider (ECSP) management system, which includes sending a request to a management and orchestration (MANO) to reserve resources based on location and performance parameters, establishing a network connection with a user plane function (UPF), and deploying the EAS after successful reservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a 'first come, first served' mechanism is used for EAS deployment, then resource allocation is simple, but deployment failure rate increases due to insufficient remaining resources
Solution Approach 1:
The patent applies preliminary action by allowing ASPs to reserve resources in advance before deploying EAS applications. The ECSP management system receives requirement information from ASPs and sends requests to MANO to reserve resources in EDNs before the actual deployment occurs. This advance reservation ensures that resources are available when needed, reducing deployment failures while maintaining manageable complexity through automated resource tracking and release mechanisms.
2Reliability
If resources are reserved in advance for EAS deployment, then deployment success rate improves, but resource utilization may decrease due to long-term occupation
Solution Approach 1:
The patent implements dynamics by making resource reservations time-bound and conditional rather than permanent. Resources are reserved for specific time periods or until deployment completion, after which they are automatically released. The system dynamically tracks reservation status, expiration times, and actual deployment outcomes, allowing reserved resources to be reallocated to other ASPs if not used within the reservation period. This dynamic approach ensures high deployment success rates while maintaining efficient resource utilization through automatic release and reallocation mechanisms.
3Speed
If EAS is deployed without pre-established network connection, then deployment process is faster, but latency increases due to connection establishment delay
Solution Approach 1:
The patent applies preliminary action by establishing network connections between EDNs and UPFs in advance during the resource reservation phase. When the ECSP management system reserves resources in an EDN, it simultaneously sets up the necessary network connectivity to the UPF. This pre-establishment of network connections means that when EAS deployment occurs, the communication pathway is already ready, eliminating connection establishment delays and reducing latency for edge computing services while maintaining efficient deployment throughput.
Data Source
AI summary
An edge application method, performed by an edge computing service provider (ECSP) management system, includes receiving requirement information of a reserved resource from an application service provider (ASP). The requirement information of the reserved resource comprises location information of the reserved resource. The edge application method also includes sending a first request to a management and orchestration (MANO) based on the requirement information of the reserved resource. The first request is used to request the MANO to reserve the resource in an edge data network (EDN) corresponding to the location information, and the reserved resource is used to deploy an edge application server (EAS) of the ASP.