Adaptive Idle Detection in Hyper-Converged Infrastructure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current idle detection techniques in virtualized computing environments are inaccurate due to reliance on absolute resource utilization thresholds, failure to account for management overhead, and excessive overhead on production environments, leading to inefficient resource management and unnecessary costs.
Innovation Solution
An adaptive idle detection algorithm using the coefficient of variation (CV) to quantify system idleness, filtering out management overhead, and employing an idleness score to reduce unnecessary data retrieval, allowing for more accurate and efficient detection of idle systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If absolute resource utilization thresholds are used for idle detection, then the detection process is simple, but the detection accuracy is poor
Solution Approach 1:
The patent transforms the idle detection approach from using absolute resource utilization thresholds to using the coefficient of variation (CV) of resource usage. This parameter change enables the system to detect idle SDDCs more accurately by measuring relative variability rather than absolute values, resolving the contradiction between detection accuracy and algorithm complexity.
Solution Approach 2:
The patent implements a dynamic idle detection mechanism that adapts to different SDDC workloads and time periods. By using rolling CV calculations and adjusting detection parameters based on observed patterns, the system maintains high accuracy without requiring overly complex static thresholds, thus resolving the contradiction between precision and complexity.
2Measurement precision
If comprehensive resource usage data is continuously monitored, then detection accuracy improves, but system overhead increases
Solution Approach 1:
The patent extracts only the essential information needed for idle detection by using the coefficient of variation calculated from resource usage data. Instead of continuously analyzing all raw resource metrics, the system extracts the CV parameter which captures the essence of workload variability, thereby maintaining detection accuracy while significantly reducing computational overhead.
Solution Approach 2:
The patent uses existing resource usage data that is already being collected for other purposes and creates a derived metric (CV) for idle detection. This approach avoids the need for additional monitoring infrastructure or excessive data collection, reducing overhead while maintaining detection precision through clever use of available data.
3Measurement precision
If management overhead is included in resource usage data, then data completeness is maintained, but idle detection accuracy deteriorates
Solution Approach 1:
The patent segments the resource usage data into different components by analyzing the coefficient of variation across various resource types. This segmentation allows the system to identify patterns specific to workload usage versus management overhead, enabling accurate idle detection even when management overhead is present in the data, thus resolving the contradiction between accuracy and information loss.
Data Source
AI summary
An adaptive idle detection method determines whether software defined data centers (SDDCs) in a hyperconverged infrastructure (HCl) environment are idle. Idleness may be quantified via a coefficient of variation (CV) against resource usage, so as to adapt the idle detection method to SDDCs with different hardware specifications and workloads. Management overhead may also be filtered out by the idle detection method, and the idle detection method may use idleness scores to further reduce overhead.