Dynamic NAND Block Configuration for Capacity and Throughput Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices with 3D NAND flash memory face issues such as reduced process yields and increased damaged storage units, leading to performance degradation and failure in storage capacity tests, as manufacturers attempt to increase capacity.
Innovation Solution
A method and apparatus for dynamic block configuration in a memory device, where blocks are divided into first and second regions, with first blocks reserved for data access and second blocks used to maximize storage capacity by adjusting reserved block thresholds, allowing for superblock configurations that balance throughput and storage capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If flash memory manufacturers enlarge three-dimensional (3D) NAND flash stack structures to increase storage capacity, then storage capacity is improved, but process yields decrease and the average number of damaged storage units increases
Solution Approach 1:
The patent segments the flash memory storage space into multiple regions with different block configuration strategies. Specifically, it divides blocks into those that can form complete superblocks and those that cannot, managing them differently to optimize both capacity utilization and reliability. This segmentation allows the system to work around damaged storage units while maximizing usable capacity.
Solution Approach 2:
The patent dynamically changes the block configuration parameters based on the actual state of storage units. It adjusts the number of blocks per superblock, the reservation of blocks for replacement, and the distribution of data across regions based on detected damaged units. This parameter adaptation enables the system to maintain optimal performance despite variations in manufacturing yield.
2Quantity of substance
If flash memory manufacturers enlarge three-dimensional (3D) NAND flash stack structures to increase storage capacity, then storage capacity is improved, but the average number of damaged storage units increases
Solution Approach 1:
The patent converts the presence of damaged storage units from a harmful factor into a manageable condition. By detecting damaged blocks and dynamically reconfiguring block assignments, the system transforms what would be dead space into opportunity for optimized resource allocation. The damaged units trigger a reconfiguration process that actually improves overall system efficiency by better matching data to functional storage space.
Solution Approach 2:
The patent performs preliminary detection and classification of damaged storage units before normal data storage operations begin. This advance identification allows the system to pre-configure block assignments that avoid damaged areas, and to reserve appropriate replacement blocks in advance. The preliminary action prevents data loss and operational failures by proactively addressing the damaged units.
3Productivity
If storage units are accessed simultaneously to improve throughput, then data access throughput is improved, but access operations may fail when damaged storage units are included in the accessed set
Solution Approach 1:
The patent implements dynamic block configuration that adapts to the actual state of storage units. Rather than using fixed superblock structures, the system continuously monitors storage unit health and reconfigures block assignments in real-time. This dynamics ensures that simultaneous access operations always target functional storage units, maintaining both high throughput and reliable access success rates.
Solution Approach 2:
The patent incorporates feedback mechanisms that monitor the state of storage units and adjust block configurations accordingly. When damaged units are detected during access operations, the system receives feedback and reconfigures the block structure to exclude problematic units. This closed-loop control maintains access reliability while preserving throughput by quickly adapting to changing storage conditions.
4Reliability
If blocks are reserved for data access to ensure throughput, then data access reliability is improved, but available storage capacity decreases
Solution Approach 1:
The patent applies partial reservation of blocks for reliability, rather than reserving excessive blocks. It dynamically determines the minimum necessary reservation based on detected damaged units and configures only that amount as reserved regions. The remaining blocks are fully utilized for data storage, achieving the optimal balance between access reliability and available capacity without over-reserving.
Solution Approach 2:
The patent dynamically changes the reservation parameter based on the actual state of storage units. When damaged units are detected, the system adjusts the number of reserved blocks upward only to the extent necessary to maintain reliability. When storage units are healthy, the reservation parameter is reduced to maximize available capacity. This parameter adaptation eliminates the need for fixed, conservative reservations.
Data Source
AI summary
A method for performing storage space management of a memory device with aid of dynamic block configuration includes: configuring at least one portion of blocks among a plurality of blocks to be multiple first blocks in a first region and multiple second blocks in a second region according to a first reserved block threshold; combining the multiple first blocks into a set of first superblocks in the first region; and combining at least one portion of second blocks among the multiple second blocks into a set of second superblocks in the second region, wherein the first reserved block threshold is less than a minimum non-damaged block count among respective non-damaged block counts of a plurality planes, for a memory controller to increase available storage capacity by increasing a ratio of a size of the second region to a size of the first region.