Virtual Machine Power Prediction Using Server Efficiency Indices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need to accurately predict the power consumption of a virtual machine when it is moved from a source server to a destination server, especially to optimize the utilization of renewable energy and reduce electricity costs.
Innovation Solution
A virtual machine power consumption prediction device that calculates the power consumption of a virtual machine on a source server based on resource usage and predicts the power consumption on a destination server using an energy consumption efficiency index for each server.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If VMs are moved to servers with surplus renewable energy supply to increase utilization rate, then renewable energy utilization is improved, but power consumption prediction accuracy deteriorates due to server-specific energy efficiency variations
Solution Approach 1:
The patent applies parameter changes by introducing an energy consumption efficiency index that characterizes each server's power consumption characteristics. This index serves as a correction parameter that adjusts the predicted power consumption based on the specific server where the VM will be migrated, thereby maintaining prediction accuracy across different servers with varying energy efficiency profiles
Solution Approach 2:
The patent uses an energy consumption efficiency index as an intermediary element between the VM migration decision and the power consumption prediction. This index acts as a mediator that translates server-specific hardware characteristics into a quantifiable factor that can be used to adjust predictions, bridging the gap between generic migration benefits and specific server performance
2Measurement precision
If power consumption prediction is performed for each server to improve accuracy, then prediction accuracy is improved, but calculation complexity increases due to multiple server comparisons
Solution Approach 1:
The patent reduces complexity by transforming the prediction problem into a parameter-based calculation. Instead of performing complex simulations or measurements for each server, the system uses a simple energy consumption efficiency index parameter that can be quickly applied to adjust base predictions, significantly reducing computational complexity while maintaining accuracy
Solution Approach 2:
The patent applies local quality by assigning specific energy consumption efficiency characteristics to each server individually. Rather than using a single average value for all servers, each server has its own indexed parameter that reflects its local hardware characteristics, allowing accurate predictions without requiring complex inter-server comparisons
Data Source
AI summary
A virtual machine power consumption prediction device includes: a power consumption calculation part which calculates power consumption of a virtual machine running on a first server on the basis of a resource usage state on the first server; and a power consumption prediction part which predicts power consumption of the virtual machine when the virtual machine is moved to a second server on the basis of an index indicating energy consumption efficiency of each of the first server and the second server.