Extended Page Patching for Storage Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer data storage devices often have unusable logical data portions due to corruption or defects, which can lead to inefficient resource consumption and system downtime during correction procedures, and may result in false failure indications where valid pages are incorrectly repaired while defective pages remain unrepaired.
Innovation Solution
An extended patching procedure is applied, where not only the explicitly defective page but also related additional pages are repaired to minimize future failure probabilities and reduce the number of repair events, by considering statistical probabilities, I/O subsystem information, and historical error data to determine the optimal list of pages for repair.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If only the explicitly defective page is repaired, then the correction procedure is simple and targeted, but the number of repair events increases and system efficiency decreases
Solution Approach 1:
The patent applies preliminary action by proactively repairing not only the defective page but also a predetermined number of additional pages (e.g., 16 pages) that may potentially be defective. This preemptive approach ensures that when pages are accessed, they are already repaired and ready, reducing the need for subsequent repair events and improving overall system efficiency.
2Productivity
If additional related pages are included in the repair procedure, then the number of repair events is reduced and efficiency is improved, but the complexity of the correction procedure increases
Solution Approach 1:
The patent applies parameter changes by modifying the scope of the repair operation from a single page to a predetermined number of additional pages. This parameter adjustment (expanding from 1 to 17 pages per repair event) balances the complexity increase with significant efficiency gains, as multiple potential defects are addressed in a single repair event.
3Productivity
If expanded read operations are performed to achieve better performance, then data retrieval efficiency is improved, but false failure indications increase causing valid pages to be incorrectly repaired
Solution Approach 1:
The patent applies the extraction principle by isolating the failure detection mechanism from the expanded read operations. The system reads expanded pages for performance optimization but uses a separate, targeted correction procedure that only repairs pages actually identified as defective through proper failure detection, preventing false repairs of valid pages while maintaining the performance benefits of expanded reads.
Data Source
AI summary
The embodiments described herein generally relate to methods and systems for using an extended patching procedure for correction or repair of logical data portions, pages, or sectors of a computer data storage device. The extended patching procedure targets for repair not only the page(s) appearing to be defective or unusable based on a failed read operation for a data transfer request, but also additional pages. Determining the additional pages to include for automatic patching is based on: statistical distribution analyzes to include pages within the physical or logical vicinity of the original page, information about the underlying storage device technology or Input/Output (I/O) subsystem, and/or historical data about error conditions for areas related to the original page. Preemptively patching pages based on extended page lists improves system performance by reducing the total number of costly repair processes and by avoiding situations involving correction actions that fail to resolve.