Redundant Storage Read Recovery Using ECC and Parity Substitution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Solid-state storage systems face challenges in maintaining data integrity due to wear-out mechanisms and errors, particularly in redundant storage systems where data failures and bit errors can occur, leading to reduced system reliability and efficiency.
Innovation Solution
A system and method that include modules for receiving read requests, reading data from a subset of storage elements, correcting errors using Error Correcting Code (ECC), and substituting data from alternate storage elements to maintain data integrity, thereby reducing the number of reads and extending the lifespan of storage elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is read frequently from storage elements to ensure data availability, then data accessibility is improved, but wear on storage elements increases and data integrity deteriorates
Solution Approach 1:
The system performs preliminary actions by reading data from alternate storage elements before actual wear occurs on primary elements. By pre-positioning data in redundant locations and using error correction codes in advance, the system can retrieve data without repeatedly accessing the same worn storage elements, thus maintaining accessibility while preserving integrity.
Solution Approach 2:
The patent implements copying by maintaining redundant copies of data across multiple storage elements. When data is needed, the system can retrieve it from a copy rather than the original worn element. Error correction codes also create virtual copies of data integrity information that can restore original data without re-reading degraded storage elements.
2Reliability
If the number of storage elements is increased to provide redundancy, then data integrity is improved, but the number of reads required increases which accelerates wear
Solution Approach 1:
The system segments the redundant storage array into distinct functional groups: primary data storage elements and alternate storage elements. This segmentation allows the system to route read operations preferentially through alternate elements during normal operations, preserving the lifespan of primary elements while maintaining the redundancy benefit of having multiple storage elements.
Solution Approach 2:
Error correction codes serve as an intermediary mechanism that reduces the need for direct re-reading of failed or worn storage elements. When a storage element degrades, the ECC system can reconstruct the original data through mathematical relationships with other storage elements, acting as a mediator that prevents direct access to worn elements and thus extends their usable life.
3Reliability
If error correction codes are applied to all stored data, then data integrity is improved, but the computational complexity and processing time increase
Solution Approach 1:
The system applies local quality by using error correction codes selectively rather than uniformly across all data. ECC is applied at specific granularities (e.g., at the storage element level or for critical data portions) rather than to every single bit of data, reducing overall computational complexity while maintaining integrity where it matters most. The patent also applies different levels of redundancy to different data regions based on their importance.
Data Source
AI summary
In various embodiments, an apparatus, system, and method may increase data integrity in a redundant storage system. In one embodiment, a request is received for data stored at a storage system having a plurality of storage elements, where one or more of the plurality of storage elements include parity information. A determination is made that one of the plurality of storage elements is unavailable, the unavailable storage element being a functional storage element and including at least a portion of the data. Responsive to the determination, the data is reconstructed based on at least a portion of the parity information and data from one or more of the plurality of storage elements other than the unavailable storage element; a response is provided to the request such that the response includes the reconstructed data.