Server Consolidation Using Magnitude Classification Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current server consolidation methods, such as the First Fit Decreasing (FFD) and Least Loaded (LL) algorithms, are limited in optimizing multiple resources simultaneously, leading to inefficiencies in determining optimal server allocation in dynamic environments with changing workloads, as they require an indeterminate amount of time and are not suited for real-time adjustments.
Innovation Solution
The proposed method employs a Magnitude Classification Model (MCM) using fuzzy logic to convert continuous resource utilization values into discrete magnitudes, generating a combination matrix, and optimizing it to determine the optimal packing of servers across multiple dimensions, allowing for simultaneous consideration of CPU, disk, I/O, and memory utilization, thereby reducing the number of destination servers needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional server consolidation algorithms (FFD, LL) are used, then server allocation can be performed, but the optimization of multiple resources simultaneously is limited and the time required is indeterminate
Solution Approach 1:
The patent transforms the continuous multi-dimensional resource allocation problem into a discrete optimization problem by defining magnitude thresholds for different resource dimensions. This parameter transformation enables the use of efficient discrete optimization algorithms while maintaining adequate optimization accuracy, resolving the contradiction between speed and precision in server consolidation.
2Quantity of substance
If the number of physical servers is minimized, then cost is reduced, but sufficient resources may not be available avoiding performance degradation
Solution Approach 1:
The patent introduces magnitude thresholds as an additional dimensional constraint to traditional resource allocation. By defining acceptable magnitude ranges for resource utilization across multiple dimensions (CPU, memory, storage, I/O), the system can determine optimal server consolidation while guaranteeing performance requirements are met, thus minimizing physical servers without compromising reliability.
3Adaptability or versatility
If existing algorithms are used for server consolidation, then some resource optimization is achieved, but they are not suited for real-time adjustments in dynamic environments
Solution Approach 1:
The patent replaces complex continuous optimization calculations with a discrete magnitude-based optimization approach. By converting continuous resource utilization values into discrete magnitude categories and using threshold-based optimization, the system achieves real-time adaptability to dynamic workload changes while keeping calculation time deterministic and suitable for real-time environments.
Data Source
AI summary
A computer-implemented method, apparatus, and non-transitory computer-readable medium for determining optimal combinations of elements having multiple dimensions, including removing all multi-dimensional elements from a combination matrix which have a dimension corresponding to a highest classification in a plurality of classifications, iteratively combining one or more multi-dimensional elements from a first end of the combination matrix and one or more multi-dimensional elements from a second end of the combination matrix to generate one or more combined multi-dimensional elements, incrementing a count of packed combinations when a combined multi-dimensional element in the one or more combined multi-dimensional elements has a dimension corresponding to the highest classification in the plurality of classifications, and removing a combined multi-dimensional element in the one or more combined multi-dimensional elements from the combination matrix when the combined multi-dimensional element has a dimension corresponding to the highest classification in the plurality of classifications.