Magnetic Disc Failure Prediction via Error Position Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current magnetic disc devices lack effective methods for predicting failures, leading to potential data loss and unnecessary replacements, as they cannot accurately differentiate between head and disc failures based on error distribution patterns.
Innovation Solution
A failure predicting method that collects error information and uses parameters like dispersive power to determine the physical position relationships between errors on the magnetic disc, distinguishing between head and disc failures by analyzing error localization and dispersal, thereby enabling early detection and targeted maintenance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If magnetic disc devices conduct immediate replacement upon detecting errors, then data loss is prevented, but unnecessary replacements increase due to inability to differentiate between head and disc failures
Solution Approach 1:
The patent segments error analysis by spatial distribution patterns, dividing errors into localized clusters (indicating disc failures) and dispersed patterns (indicating head failures). This segmentation enables differentiated response strategies: immediate replacement for localized disc failures versus monitoring for dispersed head failures, thereby preventing unnecessary replacements while maintaining data loss prevention.
Solution Approach 2:
Instead of assuming all errors indicate immediate failure requiring replacement, the patent inverts the logic by using error distribution patterns to determine when replacement is NOT needed. By analyzing whether errors are localized or dispersed, the system identifies cases where continued operation is safe, thus reducing unnecessary replacements while still preventing data loss in critical cases.
2Measurement precision
If magnetic disc devices monitor all errors continuously, then failure prediction accuracy improves, but system complexity and computational load increase
Solution Approach 1:
The patent extracts only the essential spatial distribution characteristics of errors (localized versus dispersed patterns) rather than analyzing all error parameters continuously. By focusing extraction on position relationships and clustering patterns, the system achieves accurate failure prediction while minimizing computational complexity, as it only needs to track error locations and their spatial relationships rather than all possible error attributes.
Solution Approach 2:
The patent transforms continuous error monitoring into discrete pattern classification by changing the parameter from continuous error data to categorical failure mode identification (head failure versus disc failure). This parameter change simplifies the system by converting complex continuous monitoring into discrete pattern recognition, reducing computational load while maintaining prediction accuracy.
3Reliability
If magnetic disc devices perform comprehensive error analysis, then failure mode differentiation improves, but processing time and operational overhead increase
Solution Approach 1:
The patent performs preliminary analysis of error position relationships and establishes clustering patterns in advance, creating a framework for rapid failure mode determination. By pre-establishing the spatial analysis framework and clustering criteria, the system can quickly differentiate between head and disc failures when errors occur, minimizing processing time while maintaining accurate failure mode differentiation.
Data Source
AI summary
According to one embodiment, physical position information on errors on a recording medium is acquired, physical position relationship between the errors on the recording medium is calculated based on the position information, and a failure mode related to the errors is determined based on the position relationship.