Non-intrusive Storage Performance Monitor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring storage device performance, such as benchmarking, are resource-intensive and disruptive, as they require temporarily removing drives from production use and can be time-consuming, even when the drive is performing optimally.
Innovation Solution
A non-intrusive monitoring system that collects and analyzes input/output data from storage devices, generates models based on this data, and compares the performance of individual drives to a control drive, allowing for real-time detection of sub-optimal performance without disrupting ongoing operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional benchmarking methods are used to monitor storage device performance, then performance data can be obtained, but the process is time-consuming and requires removing drives from production use
Solution Approach 1:
The system performs preliminary actions by continuously collecting I/O performance data during normal production operations before any benchmarking is needed. This allows performance baselines to be established in advance, eliminating the need to remove drives from production for monitoring purposes.
Solution Approach 2:
The patent introduces an intermediary approach by using a control storage device as a reference point. Instead of directly benchmarking each drive, the system compares individual drive performance against the control device, which has been characterized in advance, enabling indirect performance assessment without production disruption.
2Reliability
If traditional benchmarking methods are used to monitor storage device performance, then performance degradation can be detected, but drives must be temporarily removed from production use
Solution Approach 1:
The system ensures continuity of useful action by maintaining uninterrupted I/O operations to and from storage devices during the monitoring process. Data collection occurs continuously in the background without interrupting production workflows, and the control device remains available as a permanent reference.
Solution Approach 2:
The monitoring system performs self-service by automatically collecting performance data during normal operations and comparing it against the control device baseline. This automated process eliminates the need for manual benchmarking interventions that would require removing drives from production.
3Measurement precision
If comprehensive performance monitoring is implemented, then sub-optimal operation can be detected, but the monitoring process becomes complex and resource-intensive
Solution Approach 1:
The patent applies local quality by focusing monitoring efforts on specific, locally-defined performance characteristics of the control storage device. Instead of attempting to measure all possible parameters across all drives, the system identifies and monitors key performance attributes that are locally relevant to each storage device's function.
Data Source
AI summary
A method of monitoring a plurality of storage devices using a storage management system includes receiving and analyzing input/output data related to the storage devices and generating a model in response to the monitoring, including the detected input/output data related to at least a first and a second storage device, and where the input/output data includes at least one measurable storage device aspect. The method also includes determining, in response to the generating, that the first storage device is a control storage device, and that the second storage device is to be compared to the first storage device and comparing the input/output data of the second storage device to the first storage device according to the at least one measurable storage device aspect. The method also includes determining that the second storage device is operating sub-optimally and transmitting a notification that the second storage device is operating sub-optimally.