Virtual Machine Arrangement Search for Data Center Power Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods face challenges in optimizing electric power consumption in data centers due to unknown parameters for mixed-integer programming problems, varying values over time, and the inability to update parameters effectively, leading to difficulties in reducing overall power consumption.
Innovation Solution
A non-transitory computer-readable recording medium stores an arrangement search program that sets initial values for virtual machine placement based on performance information and heat coupling data, using a sequential parameter estimation method to optimize power consumption by updating parameters to advantageous values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If traditional mixed-integer programming methods are used to optimize power consumption, then power consumption optimization is attempted, but parameters are unknown and cannot be updated effectively
Solution Approach 1:
The patent applies preliminary action by setting initial values for parameters before solving the mixed-integer programming problem. The system determines initial parameter values based on performance information and heat coupling data obtained in advance, allowing the optimization process to start with informed estimates rather than unknown parameters.
Solution Approach 2:
The patent implements feedback through the sequential parameter estimation method, which updates parameter values based on observed performance data. The system continuously refines parameter estimates by comparing predicted power consumption with actual measurements, allowing parameters to be updated effectively over time based on facility performance feedback.
2Measurement precision
If parameters are updated using sequential parameter estimation method, then parameter accuracy improves, but time for parameter convergence increases
Solution Approach 1:
The patent reduces convergence time by performing preliminary determination of initial parameter values using performance information and heat coupling data. This preliminary action provides a starting point that is already close to optimal values, significantly reducing the number of iterations needed for the sequential parameter estimation method to converge.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting parameter values through the sequential estimation process. The system changes parameters from initial estimates to refined values based on observed performance, allowing accurate parameter determination while minimizing convergence time through efficient update strategies.
3Use of energy by moving object
If optimal parameters are set based on performance information and heat coupling data, then power consumption is reduced, but system complexity increases
Solution Approach 1:
The patent applies universality by creating a unified optimization system that handles multiple functions: collecting performance information, determining heat coupling data, setting initial parameter values, solving mixed-integer programming problems, and updating parameters. This multi-functional approach integrates various tasks into a single coherent system rather than separate complex components.
Solution Approach 2:
The system applies self-service by automatically determining initial parameter values and updating parameters through the sequential estimation method without requiring manual intervention. The system uses its own performance data and heat coupling information to self-optimize, reducing the need for external complexity while achieving power consumption reduction.
Data Source
AI summary
A non-transitory, computer-readable recording medium stores therein an arrangement search program that causes a computer that searches arrangement of virtual machines in plural servers in a facility including the plural servers to execute a process that includes setting an initial value of a parameter concerning the arrangement of the plurality of virtual machines in the plurality of servers, based on at least any one of first performance information on power consumption of the plurality of servers, second performance information on power consumption of air conditioning equipment installed in the facility, third performance information on power consumption of power source equipment installed in the facility, and heat coupling information on heat coupling among the plurality of servers and among the plurality of servers and the air conditioning equipment; and updating the parameter by a sequential parameter estimation method, so as to optimize power consumption of the overall facility.