Virtual Machine Allocation Using Predicted Renewable Power Surplus
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for allocating virtual machines across bases with renewable energy power generation do not effectively consider surplus power, leading to potential waste of renewable energy.
Innovation Solution
A virtual machine allocation device that predicts power generation and consumption at each base, selects time periods with surplus power, and allocates movable virtual machines to optimize energy usage, thereby minimizing surplus power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If virtual machines are allocated without considering surplus power of renewable energy, then allocation simplicity is maintained, but renewable energy waste occurs
Solution Approach 1:
The system performs preliminary prediction of renewable energy power generation amounts and power consumption for each base before VM allocation. By predicting surplus power in advance and incorporating this information into the allocation decision-making process, the system proactively prevents renewable energy waste rather than reacting to it after the fact.
Solution Approach 2:
The allocation unit uses predicted surplus power information as feedback to dynamically adjust VM allocation decisions. The system continuously monitors and responds to changes in renewable energy availability by reallocating VMs to bases with surplus power, creating a closed-loop control system that adapts to varying energy conditions.
2Productivity
If virtual machines are frequently moved between bases to optimize renewable energy usage, then energy efficiency improves, but system operational complexity increases
Solution Approach 1:
The system predicts power generation and consumption patterns in advance, allowing VM allocation decisions to be made based on forecasted conditions rather than reactive adjustments. This preliminary planning reduces the frequency and complexity of VM movements by anticipating optimal allocation scenarios before they are needed.
3Adaptability or versatility
If renewable energy power generation fluctuates, then flexibility in power supply is maintained, but VM allocation stability deteriorates
Solution Approach 1:
By predicting renewable energy power generation amounts and power consumption in advance, the system prepares allocation plans that anticipate fluctuations in power supply. This allows the system to maintain VM allocation stability by having pre-planned responses to expected variations in renewable energy availability.
Solution Approach 2:
The system dynamically adjusts VM allocations based on predicted surplus power conditions while maintaining overall allocation stability through systematic decision-making. The allocation unit responds to changing energy conditions in a controlled manner, balancing adaptability to power fluctuations with stability in VM service delivery.
Data Source
AI summary
A virtual machine allocation apparatus configured to perform allocation of virtual machines to a plurality of bases to which supply of electric power is performed by renewable energy power generation, includes: a prediction unit configured to acquire, for each base, a predicted value of an amount of power generation of renewable energy power generation and a predicted value of power consumption; and an allocation unit configured to repeatedly execute processing of selecting a base and a time period in which surplus power of the renewable energy power generation is maximized within a predetermined time length divided into a plurality of time periods at the plurality of bases, allocating a virtual machine selected from a control target virtual machine group including one or more movable virtual machines to the selected base and time period, and excluding the allocated virtual machine from the control target virtual machine group.