SSD Controller Data Duplication Parity Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face challenges in efficiently duplicating data across multiple solid state disks (SSDs) for redundancy, particularly in ensuring data integrity and recovery in case of SSD failures, and optimizing storage space usage.
Innovation Solution
A data storage device configuration that utilizes a controller to perform primary and secondary data duplication across multiple SSDs, employing first-type memory for primary duplication and second-type memory for secondary duplication, with parity data generated and stored on a parity SSD, allowing for efficient data recovery and optimized storage density.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is duplicated across multiple SSDs using traditional methods, then data redundancy and reliability are improved, but storage space efficiency deteriorates due to complete duplication of all data segments
Solution Approach 1:
The data is divided into multiple segments that are distributed across different SSDs. Instead of duplicating entire data sets, only specific data segments are replicated on target SSDs, while parity segments are stored on parity SSDs. This segmentation approach maintains data redundancy while optimizing storage space utilization.
Solution Approach 2:
The system changes the parameter of data representation by storing parity information instead of complete data copies. Parity segments are generated from source data segments using XOR operations, transforming the storage approach from full duplication to differential storage, thereby improving space efficiency while maintaining reliability.
2Speed
If parity data is stored in first-type memory (NVRAM), then data recovery speed is improved, but storage capacity is reduced compared to using second-type memory
Solution Approach 1:
Different types of memory are assigned different functions based on their characteristics. First-type memory (NVRAM) with faster access speed is used specifically for storing parity segments that require quick access during data recovery. Second-type memory (NAND flash) with larger capacity is used for storing source data segments. This local quality differentiation optimizes both recovery speed and storage capacity.
Solution Approach 2:
The system introduces a dimensional distinction in memory hierarchy by separating fast-access NVRAM from high-capacity NAND flash. Parity data is placed in the fast-access dimension (NVRAM) to accelerate recovery operations, while bulk data resides in the high-capacity dimension (NAND flash), creating a multi-dimensional storage architecture.
Data Source
AI summary
A data storage device includes a plurality of solid state disks (SSDs) and a controller configured to control the SSDs. The controller respectively writes a first data segment and duplicated data of the first data segment to first-type memories of two respective SSDs among the plurality of SSDs in response to a first write request of a host, respectively writes a second data segment and duplicated data of the second data segment to first-type memories of two respective SSDs among the plurality of SSDs in response to a second write request of the host, and writes parity data generated based on the duplicated data to a first-type memory or a second-type memory of a parity SSD. The data storage device may be included in a data processing system.