NAND Block Pool Reallocation for Higher TBW and Wear Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing NAND based storage devices face challenges in accurately estimating and managing the usage of block pools, leading to uneven wear and reduced Terabytes Written (TBW) due to unpredictable workloads and data types, resulting in premature end-of-life scenarios.
Innovation Solution
A block reallocating system monitors the Greatest Erase Count (GEC) of independent block pools and reallocates free blocks from less used pools to those approaching maximum erase counts, ensuring dynamic and efficient utilization of storage resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static Over Provisioning is used for different block pools, then data types are separated and managed independently, but TBW is reduced due to uneven wear and premature block pool exhaustion
Solution Approach 1:
The patent implements dynamic Over Provisioning adjustment by monitoring GEC of block pools and reallocating free blocks from less worn pools to more worn pools. The system calculates reallocation ratios based on GEC differences and transfers blocks accordingly, transforming the static OP configuration into a dynamic adaptive mechanism that extends TBW while maintaining data type separation.
2Productivity
If dynamic block reallocation is implemented, then TBW is extended by balancing wear across block pools, but device complexity increases due to monitoring and reallocation mechanisms
Solution Approach 1:
The system continuously monitors GEC of block pools and uses this feedback to determine reallocation needs. When GEC difference between pools exceeds a threshold, the system calculates reallocation ratios and executes block transfers. This closed-loop feedback mechanism enables automatic wear balancing without complex external intervention.
Solution Approach 2:
The block reallocation system operates autonomously by self-monitoring GEC, self-determining reallocation needs, and self-executing block transfers. The system manages its own wear balancing without requiring external control, reducing operational complexity while extending TBW.
Data Source
AI summary
The present disclosure relates to the field of storage devices and provides methods and systems for improving Terabytes Written (TBW) of a NAND based storage device. A block reallocating system monitors a Greatest Erase Count (GEC) of each independent block pool at intervals. Further, the block reallocating system determines a candidate independent block pool having a rate of increase of Erase Count (EC) higher than a rate of increase of EC of remaining ones of the plurality of independent block pools, based on the GEC. Further, it is determined whether the GEC of the candidate independent block pool and at least one independent block pool meet a reallocation condition, and then one or more free blocks are reallocated from the at least one independent block pool to the candidate independent block pool for improving TBW of the NAND based storage device.