Memory Controller Super Block Classification for Bad Block Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face inefficiencies in managing and utilizing memory blocks, particularly when dealing with a mix of good and bad memory blocks, leading to reduced reliability and storage capacity due to the inability to effectively group and utilize all memory blocks within a super memory block framework.
Innovation Solution
A memory system and method that groups memory blocks into super blocks, classifying them based on the presence of bad blocks, and strategically distributes write data across different types of super blocks to optimize storage and utilization, using a controller to manage the distribution and reassignment of memory blocks to maintain efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If memory blocks are grouped into super blocks with strict separation of good and bad blocks, then reliability is improved, but storage capacity and utilization deteriorate
Solution Approach 1:
The patent applies local quality by differentiating the treatment of good blocks and bad blocks within super blocks. Good super blocks contain only good memory blocks for reliable storage, while insufficient super blocks contain a mix of good and bad blocks. The controller selectively assigns data to appropriate super block types based on data importance and super block status, optimizing both reliability and utilization.
Solution Approach 2:
The patent converts the harmful presence of bad memory blocks into a beneficial resource by creating insufficient super blocks that can still store data. Instead of completely discarding super blocks with bad blocks, the system utilizes them for storing less critical data, thereby increasing overall storage capacity while maintaining reliability for important data.
2Reliability
If bad memory blocks are excluded from super blocks, then data integrity is improved, but memory block utilization deteriorates
Solution Approach 1:
The patent segments super blocks into two categories: good super blocks containing only good memory blocks for high-integrity data, and insufficient super blocks containing mixed good and bad blocks for less critical data. This segmentation allows the system to maintain data integrity for important information while utilizing all available memory resources.
Solution Approach 2:
The patent changes the parameter of super block composition from a binary state (all good or all bad) to a graduated system where insufficient super blocks can contain a mixture of good and bad blocks. The controller dynamically adjusts which super blocks are used based on data requirements and super block status, optimizing both integrity and utilization.
3Quantity of substance
If all memory blocks are utilized including those with bad blocks, then storage capacity is improved, but system reliability deteriorates
Solution Approach 1:
The patent applies local quality by assigning different reliability levels to different super blocks based on their composition. Good super blocks provide high reliability for critical data, while insufficient super blocks provide lower reliability for non-critical data. This differentiated approach allows the system to maximize storage capacity while maintaining appropriate reliability for each data type.
Data Source
AI summary
a memory system may include: a memory device including a plurality of memory blocks; and a controller configured to manage the plurality of memory blocks as a plurality of super blocks, the controller may classify and may manage super blocks formed by mixing and grouping at least one bad memory block and normal memory blocks as first super blocks, and may classify and may manage super blocks formed by grouping only normal memory blocks as second super blocks, the controller may check an accumulated size of write data received from a host, may group the write data into a plurality of data groups based on a result of the checking of the accumulated size, and may store, each time one data group is formed, the formed one data group in N first super blocks and M second super blocks.