Virtual Machine Deployment Policy for NFVO Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for deploying virtual machines in virtual core networks face issues with slow service processing speed and poor user experience due to long scaling-out times, as they require processing time for user migration and data synchronization, leading to situations where users cannot access services promptly.
Innovation Solution
A virtual machine deployment method and device that utilize a deployment policy to elastically scale out or scale in virtual machines within a reserved range, with notifications through a Virtualized Network Function Manager (VNFM), ensuring that the number of virtual machines is maintained within a minimum and maximum threshold, and restrictions are cleared when the deployment period ends, thus avoiding service disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If virtual machines are scaled out or ejected in the related art, then service processing capacity is enhanced or reduced, but scaling-out time is long and user experience is reduced
Solution Approach 1:
The patent applies preliminary action by pre-calculating the number of virtual machines to be scaled out or in based on forecasted core network element loads before actual scaling occurs. The NFVO calculates the number of virtual machines in advance and notifies the VNFM, allowing the system to prepare scaling actions beforehand rather than reacting after load changes are detected, thereby reducing scaling-out time while maintaining service processing capacity adjustments.
2Productivity
If virtual machines are scaled out or ejected, then service processing capacity is adjusted, but time spent in user migration and data synchronization is longer
Solution Approach 1:
The patent performs preliminary calculation of the number of virtual machines to be scaled based on forecasted loads, allowing the system to plan and execute scaling actions before actual load changes occur. This advance planning reduces the time needed for user migration and data synchronization by avoiding reactive scaling operations that would trigger these time-consuming processes at suboptimal moments.
Solution Approach 2:
The patent implements a feedback mechanism where the NFVO acquires data representing current core network element loads, calculates the number of virtual machines to be scaled, and notifies the VNFM accordingly. This closed-loop feedback system ensures that scaling decisions are based on real-time load conditions, optimizing the timing of user migration and data synchronization to minimize service disruption.
3Reliability
If original virtual machines cannot process services in time, then users cannot access services, but in the related art no effective solution has been provided
Solution Approach 1:
The patent applies preliminary action by pre-calculating and preparing virtual machine scaling actions based on forecasted loads before actual service demand changes occur. This ensures that virtual machines are ready to process services without delay, maintaining both service accessibility and processing speed by avoiding the scenario where original virtual machines cannot process services in time.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
Provided are a virtual machine deployment method and device and a network function virtualization orchestrator (NFVO). Herein, the virtual machine deployment method includes: determining that start time of reserved deployment for virtual machines in a set deployment policy is reached, notifying a virtual machine management module for a virtual core network that the reserved deployment for virtual machines takes effect, herein, under a situation that the reserved deployment for virtual machines takes effect, the virtual machine management module for the virtual core network elastically scales out or scales in a number of virtual machines to be within a range of a number of reserved virtual machines (S102); determining a current elastically scaled-out or scaled-in number of virtual machines according to the deployment policy (S104); and deploying virtual machines according to the determined elastically scaled-out or scaled-in number of virtual machines (S106). By adopting the method, the problems of slow service processing speed and poor user experience caused by long scaling-out of virtual machines in the related art are solved, and thus the effects of reducing the scaling-out of virtual machines, improving the service processing speed and improving the user experience are achieved.