Storage Device Memory Segmentation for Reliability Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage devices and systems face limited lifespan due to repeated use, leading to reduced data storage reliability and increased maintenance costs, as mechanical components deteriorate and flash memory cells degrade with program and erase operations.
Innovation Solution
A storage device with multiple memory areas of varying reliability levels, controlled by a memory controller to store data based on required reliability and retention time, ensuring high-reliability data is stored in high-reliability areas for extended periods, and less critical data in lower-reliability areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is stored in memory areas with uniform reliability levels, then storage management is simple, but data reliability requirements cannot be differentiated
Solution Approach 1:
The storage device divides memory areas into multiple segments with different reliability levels (first memory area with higher reliability, second memory area with lower reliability). This segmentation allows different types of data to be stored in appropriate memory areas based on their reliability requirements, thereby improving data reliability without requiring complete uniformity across the entire storage system.
Solution Approach 2:
Different memory areas are assigned different reliability characteristics locally. The first memory area is designed with higher reliability properties while the second memory area has lower reliability properties. This local differentiation enables the storage system to match data reliability requirements with appropriate storage locations, improving overall system reliability for critical data without uniformly increasing complexity across all storage areas.
2Duration of action of moving object
If storage devices are used repeatedly to extend service time, then productivity increases, but reliability deteriorates due to wear
Solution Approach 1:
The storage device proactively monitors wear levels of memory areas and performs data migration before actual failures occur. When a memory area approaches its wear limit, data is preemptively transferred to a more reliable memory area, preventing reliability deterioration and extending the overall service time of the storage device.
Solution Approach 2:
The system dynamically changes the reliability parameters of memory areas based on wear levels. As memory areas undergo program-erase cycles, their reliability parameters degrade, and the system responds by migrating data from degraded areas to healthier areas, thereby maintaining overall storage reliability while extending service time.
3Reliability
If all data is stored with high reliability standards, then data integrity is maximized, but storage capacity is reduced
Solution Approach 1:
Instead of applying high reliability standards to all data uniformly, the system applies high reliability measures only to data that requires it (first data stored in the first memory area). Less critical data (second data) is stored in the second memory area with lower reliability standards, thereby maintaining data integrity for important data while preserving overall storage capacity.
Solution Approach 2:
Different quality levels of data integrity are applied locally to different data types. Critical data receives high reliability treatment in the first memory area, while non-critical data is stored in the second memory area with standard reliability measures. This local differentiation maximizes data integrity where needed without sacrificing overall storage capacity.
Data Source
AI summary
A storage device communicating with a host includes a plurality of memory devices and a memory controller. Each of the memory devices includes at least one of a plurality of memory areas that have different storage reliability levels. The memory controller controls the memory devices such that data and required level data associated with a required reliability level of the data are stored in some or all of the memory areas. The data and the required level data are provided from the host. The data is stored in a memory area having a storage reliability level corresponding to the required reliability level from among the memory areas, according to a control of the memory controller.