Proactive Disk Failure Prediction via Threshold Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage management systems often operate reactively, leading to performance degradation or failure due to unpredictable hard disk failures, which are sudden and without warning, and require increasing human resources for monitoring and maintenance.
Innovation Solution
A storage management computing device that proactively predicts disk failure by obtaining performance data, setting threshold values, and executing operational test cases to differentiate between actual and false failures, thereby preventing errors and improving user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reactive monitoring is used to detect disk failures, then disk failures can be detected, but performance degradation or failure has already occurred and human resources are required for monitoring
Solution Approach 1:
The system performs preliminary actions by proactively predicting disk failures before they occur. It collects performance data continuously, establishes baseline thresholds, and identifies trends indicating potential failures. This allows the system to alert administrators in advance, enabling preventive replacement of disks before actual failure impacts system operation, thus resolving the contradiction between detecting failures and avoiding performance degradation.
2Reliability
If proactive prediction is implemented, then false positives may occur, but system reliability is improved
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring disk performance data, comparing it against established thresholds, and adjusting predictions based on trends. When potential failures are predicted, the system can verify through additional monitoring or operational test cases before confirming, allowing administrators to take corrective action. This feedback loop improves reliability while managing false positives through iterative verification.
3Reliability
If manual monitoring of storage systems is performed, then disk failures can be detected, but increasing human resources are required
Solution Approach 1:
The system applies self-service by automatically collecting performance data from storage disks, analyzing the data against established thresholds, and generating failure predictions without human intervention. The automated monitoring and analysis processes eliminate the need for manual checking while maintaining high reliability in failure detection, thus resolving the contradiction between detection capability and system complexity.
Data Source
AI summary
A method, non-transitory computer readable medium, and device that assists with identifying failure of a storage disk includes obtaining performance data and a threshold value associated with each of the obtained performance data for a plurality of storage disks. Next, the obtained performance data is compared with the corresponding threshold value associated with each of the obtained performance data to determine for one or more initial failures in the plurality of storage disks. Next, one or more operational test cases are executed on each of the determined one or more initial failures in the plurality of storage disks to determine for one or more false positives in the determined one or more initial failures in the plurality of storage disks. A notification is sent based on the determination of the one or more false positives.