VM Resource Prediction for Web Conference Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing auto-scaling technologies struggle to proactively control resources, particularly in web conference services, leading to potential resource depletion and user experience quality deterioration during simultaneous high-demand events.
Innovation Solution
A resource determination device and method that utilize a model to predict the quality of service based on processing loads and resource arrangements in virtual machines, allowing for proactive adjustments to resource arrangements to meet quality requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If auto-scaling is applied based on real-time state of VM, then resource adjustment can be performed dynamically, but resources cannot be proactively arranged and user experience quality deteriorates during sudden high processing load
Solution Approach 1:
The system performs preliminary actions by predicting future processing loads and proactively adjusting resources before the actual load occurs. The prediction unit forecasts processing loads at future time points, and the control unit adjusts resources in advance based on these predictions, ensuring user experience quality is maintained during sudden high-demand events.
Solution Approach 2:
The system implements feedback by continuously monitoring actual processing loads, comparing them with predicted values, and using this information to improve future predictions and resource allocation decisions. The prediction unit is updated based on the difference between actual and predicted processing loads, creating a closed-loop control system.
2Productivity
If multiple web conferences occur simultaneously, then service coverage increases, but processing load increases causing resource depletion and quality deterioration
Solution Approach 1:
The system proactively predicts processing loads for future time points when multiple web conferences are scheduled to occur simultaneously. By predicting these peak loads in advance, the control unit can allocate additional resources before the conferences start, preventing resource depletion and maintaining service quality during high-demand periods.
Solution Approach 2:
The system dynamically changes resource allocation parameters based on predicted processing loads. When the prediction unit forecasts high processing loads due to multiple simultaneous conferences, the control unit adjusts resource parameters such as CPU allocation, memory allocation, or instance scaling to match the anticipated demand, thereby maintaining service capacity without permanent over-provisioning.
3Reliability
If resources are increased to handle peak load, then user experience quality is maintained, but resource waste occurs during low-demand periods
Solution Approach 1:
Instead of permanently increasing resources to handle peak load, the system performs preliminary predictions of processing loads and only allocates additional resources when and where they are actually needed. This on-demand proactive allocation maintains user experience quality during peak periods while avoiding resource waste during low-demand periods.
Solution Approach 2:
The system implements dynamic resource allocation where resource parameters are continuously adjusted based on predicted processing loads. Resources are scaled up proactively when high demand is predicted and scaled down when demand decreases, creating a flexible resource management system that adapts to changing conditions without permanent over-provisioning.
Data Source
AI summary
A resource determination device according to one embodiment includes: a model storage device that stores a model indicating a relationship among a processing load arranged in a virtual machine, resource arrangement of the virtual machine, and a predicted value of quality when the processing load is arranged in the virtual machine; a determination unit that determines whether or not a predicted value of the quality when it is assumed that a new processing load is arranged in the virtual machine in the model satisfies an appropriate quality requirement; and a control unit that controls change of resource arrangement of the virtual machine such that the predicted value satisfies the appropriate quality requirement when the determination unit determines that the predicted value does not satisfy the appropriate quality requirement.