Dynamic NAND Aggregation Size Adjustment for Wear Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing NAND storage systems experience performance degradation and quality of service issues due to worn pages in programmable blocks, which lead to increased logic overhead and inefficient aggregation strategies.
Innovation Solution
Implementing a bad block prediction unit in the controller to dynamically adjust the aggregation size of programmable blocks based on the life cycle of the block and the number of reconstructions, allowing for a switch from block-sized to page-sized aggregation when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If block-sized aggregation is used for data writing, then write efficiency is improved, but performance degrades when bad pages are present due to increased logic overhead and reconstruction needs
Solution Approach 1:
The patent implements dynamic aggregation size adjustment based on NAND wear level. The controller monitors the number of bad pages in programmable blocks and automatically switches between block-sized aggregation (when healthy) and page-sized aggregation (when degraded), allowing the system to adapt its aggregation strategy to current device conditions rather than using a fixed approach
Solution Approach 2:
The patent changes the aggregation size parameter from a fixed block size to a variable parameter that can be adjusted between block size and page size based on the wear level of the NAND device. This parameter change allows the system to optimize write efficiency when the device is healthy while maintaining performance when bad pages are present
2Reliability
If aggregation size is reduced to page size to handle bad pages, then performance with worn blocks is improved, but write efficiency decreases due to increased logic overhead
Solution Approach 1:
The system dynamically adjusts aggregation size based on the wear level of the NAND device. When the device is healthy with few bad pages, block-sized aggregation is used for optimal write efficiency. When the device degrades with more bad pages, the system transitions to page-sized aggregation to maintain performance, avoiding the need to always use the more efficient but less reliable block-sized aggregation
3Reliability
If worst-case scenario preparations are made for bad blocks, then reliability is improved, but power consumption increases due to unnecessary logic overhead
Solution Approach 1:
The patent implements a wear-level-based dynamic aggregation strategy that monitors the actual state of the NAND device and adjusts aggregation size accordingly. This avoids the need for worst-case scenario preparations (always using page-sized aggregation) while still providing reliable handling of bad pages when they occur, thereby reducing unnecessary power consumption associated with always-active bad page handling logic
Solution Approach 2:
The controller monitors its own wear level and automatically adjusts aggregation size based on detected bad pages, eliminating the need for external intervention or conservative worst-case configurations. The system serves itself by adapting to its own degradation state, optimizing both reliability and power efficiency
Data Source
AI summary
Instead of using programmable block size aggregation, a lower multiple of page, and down to a page size aggregation is used. A bad block prediction unit in a controller is able to predict when a programmable block has a bad page. The bad block prediction unit can lower the aggregation size of a programmable block by monitoring the life cycle of the programmable block through bad block statistic collection. When the accumulation size passes a threshold, the bad block prediction unit lowers the aggregation size. The bad block prediction unit can also predict when to lower aggregation size based on the number of reconstructions. An aggregate size level is set at a page boundary, and once the number of reconstructions reaches that page boundary, the bad block prediction unit lowers the aggregation size to page aggregation. The bad block prediction unit is able to predict both life cycle threshold changes and reconstructions changes.