VM Resource Allocation Using Priority Categories and Quality Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining resource allocation in web meetings face significant calculation challenges due to the exponential increase in combinations as the number of meetings grows, making real-time optimization impractical.
Innovation Solution
A resource determination device and method that categorize processing loads into priority-based bins, using a model to predict user experience quality, and adjust resource allocation based on these categories to reduce calculation complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all combinations of web meetings are calculated to determine optimal resource allocation, then user experience quality requirements can be satisfied, but calculation time increases exponentially as the number of web meetings increases
Solution Approach 1:
The patent segments the set of all possible web meeting combinations into multiple categories based on resource allocation patterns. Instead of evaluating all combinations exhaustively, the system divides them into groups (e.g., based on number of instances, resource distribution patterns) and evaluates representative combinations from each category, significantly reducing calculation time while maintaining quality satisfaction.
Solution Approach 2:
The patent changes the evaluation parameters by introducing category-based grouping criteria. Rather than treating all combinations uniformly, it applies different evaluation parameters to different categories of combinations, allowing efficient filtering and selection of optimal allocations without exhaustive search of all possibilities.
2Productivity
If the number of web meeting combinations is reduced through categorization, then calculation time is reduced, but the ability to satisfy user experience quality requirements may be compromised
Solution Approach 1:
The patent performs preliminary categorization of web meeting combinations before full evaluation. By pre-grouping combinations into categories based on structural characteristics, the system prepares the evaluation process in advance, allowing efficient selection of representative combinations that are most likely to satisfy user experience quality requirements without compromising thoroughness.
Solution Approach 2:
The categorization system is designed to self-adjust and self-optimize. The evaluation process automatically identifies which categories contain combinations most likely to satisfy quality requirements and focuses computational resources on those categories, allowing the system to serve itself by directing effort where it is most needed without external intervention.
Data Source
AI summary
According to a resource determination device in an embodiment, categorization information indicating that a plurality of processing loads with similar sizes, arrangement of which in virtual machines has been requested, are categorized into each of a plurality of categories to which priority orders have been applied is stored, a processing load categorized in any of the categories is extracted in accordance with the priority orders, and when a predicted value of quality when it is assumed that the processing load has been arranged in the virtual machine does not satisfy a requirement for appropriate quality, a processing load categorized into a same category as a category into which the processing load used for the determination belongs is not extracted, a processing load categorized into a category with a lower priority order or the same category is extracted, and whether or not the predicted value satisfies the requirement is determined again.