Server Utilization Factor Calculation Using Component Work and Cost Weights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining server utilization focus on usage metrics, which are inadequate as they do not account for the actual capacity and efficiency of server components, leading to inaccurate assessments of server and server-group performance.
Innovation Solution
The proposed solution calculates server and server-group utilization by deriving factors based on work and work capacity of components like processors, disks, and networking cards, and weights them by dollar costs or power usage, providing a more comprehensive evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current usage metrics methods are used to determine server utilization, then the assessment process is simple, but the accuracy of server performance assessment is inadequate
Solution Approach 1:
The patent segments server utilization assessment into multiple component-level utilization factors (CPU, disk, memory, network) rather than using a single aggregate metric. Each component's work and work capacity are calculated separately, then combined through weighted averaging to produce an overall server utilization factor, improving measurement precision through detailed component analysis.
Solution Approach 2:
The patent introduces work and work capacity as intermediary concepts to bridge the gap between simple usage metrics and accurate performance assessment. By calculating the ratio of actual work performed to maximum work capacity for each component, the system creates a standardized intermediary metric that can be accurately compared and weighted across different server components.
2Measurement precision
If component-level work and capacity calculations are implemented, then the accuracy of utilization assessment is improved, but the computational complexity increases
Solution Approach 1:
The patent changes the parameters of measurement from simple usage counts to work-based metrics that account for component capacity and efficiency. By transforming raw usage data into work ratios (actual work / maximum capacity) for each component, the system achieves more accurate measurements while maintaining computational feasibility through standardized parameter transformations.
3Adaptability or versatility
If weighting factors based on dollar cost or power usage are applied, then the comprehensiveness of evaluation is improved, but the complexity of the calculation system increases
Solution Approach 1:
The patent creates a universal weighted utilization calculation framework that can accommodate multiple weighting schemes (dollar cost, power usage, or custom weights) without requiring separate calculation systems. The same core structure of summing weighted component utilization factors can be applied with different weight assignments, providing versatility while maintaining calculation simplicity.
Data Source
AI summary
Ways are disclosed for determining a utilization of a server or group of servers. In one embodiment, a method includes determining a disk utilization factor (DUF) that indicates a utilization of one or more mass-storage disks of some server; determining a processor utilization factor (PUF) that indicates a utilization of one or more processors running on the first server; and based on the DUF and PUF deriving a server-utilization factor (SUF) that indicates a level of utilization of the first server, which includes incorporating one or more of (1) power-usage data or (2) financial-cost data that are associated with each of the disks and processors.