Coordinated Virtual Machine Autoscaling via Container Load Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In hybrid virtualization systems, autoscaling at the virtual-machine level and container level are performed independently, lacking a coordinated approach to efficiently manage resource allocation and scalability.
Innovation Solution
An information processing system that includes a container scaling apparatus, a virtual-machine scaling apparatus, a calculating unit, and a reflecting unit, which calculates and reflects the necessary number of virtual machines based on container load conditions to synchronize autoscaling across both levels, with an optional buffer ratio for resource margining.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If autoscaling at virtual-machine level and container level are performed independently, then each level can be managed separately with simple control logic, but resource allocation efficiency deteriorates and coordinated scalability is lost
Solution Approach 1:
The patent merges the previously independent virtual-machine level autoscaling and container level autoscaling into a unified coordinated autoscaling system. The container scaling apparatus and virtual-machine scaling apparatus are integrated through a calculating unit that determines the necessary number of virtual machines based on container scaling requirements, enabling joint optimization of resource allocation across both virtualization layers while maintaining manageable system complexity through modular architecture.
2Productivity
If the necessary number of virtual machines is calculated precisely based on container load, then resource allocation optimization is achieved, but calculation complexity and response time increase
Solution Approach 1:
The system performs preliminary calculations of the necessary number of virtual machines by the calculating unit based on projected container scaling requirements. This advance calculation allows the reflecting unit to pre-determine appropriate virtual machine allocation before actual container scaling occurs, reducing real-time computation burden and enabling faster response to load changes while maintaining optimization accuracy.
Solution Approach 2:
The reflecting unit implements a feedback mechanism that continuously monitors container scaling requirements and adjusts the necessary number of virtual machines accordingly. This closed-loop feedback system enables dynamic optimization of resource allocation based on actual container load conditions, achieving precise resource matching while maintaining responsive adaptation through iterative adjustments rather than complex one-time calculations.
Data Source
AI summary
An information processing system includes one or more virtual machines, a container scaling apparatus, a virtual-machine scaling apparatus, a calculating unit, and a reflecting unit. The container scaling apparatus performs autoscaling of a container that runs on the one or more virtual machines. The virtual-machine scaling apparatus performs autoscaling of the virtual machines in accordance with a difference in a necessary number of virtual machines and a number of virtual machines that are currently running. The calculating unit calculates, from a necessary number of containers calculated by the container scaling apparatus, a number of virtual machines that is necessary to cause the necessary number of containers to run thereon. The reflecting unit reflects the number calculated by the calculating unit to the necessary number of virtual machines to be used by the virtual-machine scaling apparatus.