RAID Data Protection Architecture Using DPI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data protection architectures, such as the SCSI protocol, do not effectively detect errors in RAID systems, particularly for parity data, which is crucial for data storage systems like RAID 6 that use multiple parity data sets.
Innovation Solution
A RAID structure is implemented with defined data protection information (DPI) for both payload and parity data, utilizing XOR operations and Galois Field algebra to rebuild data and protect against errors, ensuring reliable data transmission by incorporating additional data protection fields within the RAID system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If CRC error detection technology is used for data transmission, then bit errors caused by hardware signal interference can be detected, but software-caused data displacement errors cannot be detected
Solution Approach 1:
The data protection field is segmented into multiple subfields: a first subfield for storing check data of the payload data, a second subfield for storing check data of the parity data, and a third subfield for storing reference tag data. This segmentation allows different types of errors (bit errors in payload and displacement errors in parity) to be detected independently using appropriate check codes for each segment.
Solution Approach 2:
The patent introduces an intermediary data protection field between the payload data and parity data that contains multiple subfields with different check codes. This intermediary structure mediates between the payload data and parity data, enabling detection of both hardware signal interference errors in payload data and software displacement errors in parity data through the different check code mechanisms stored in the subfields.
2Reliability
If multiple parity data sets are used in RAID systems, then data protection and recovery capability are improved, but the complexity of the RAID structure increases
Solution Approach 1:
The patent merges the check data storage function into the existing data protection field structure by adding subfields within it. Instead of creating separate structures for parity data protection, the solution combines multiple check data subfields (first subfield for payload, second subfield for parity, third subfield for reference tags) within a single integrated data protection field, thereby reducing overall structural complexity while maintaining multiple parity data set protection capabilities.
3Reliability
If data protection fields are added to protect parity data, then error detection for parity data is improved, but the data transmission time increases
Solution Approach 1:
The check data for parity data is calculated and stored in the data protection field during the initial data writing process. This preliminary action of pre-calculating and storing check codes eliminates the need for time-consuming real-time verification operations during data transmission, as the protection mechanisms are already in place before errors can occur.
Data Source
AI summary
A structure of redundant array of independent disks (RAID) comprising multiple parity data is provided. A data protection field is attached after each basic data access unit of the parity data sequences and each subfields of the data protection field is defined according to different applications to protect the basic data access unit of the parity data or the data protection field of the payload data from errors incurring during data transmission.