Flash Memory Wear Balancing via Cumulative Erase Count

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In systems with multiple SSDs, some SSDs are subjected to excessive erasing and writing, leading to premature damage and potential data loss, as existing technologies lack effective high availability solutions for balancing erasing and writing operations across all storage units.

Innovation Solution

A method and device that utilize a central controller to identify and write data into the SSD with the lowest total cumulative erasing count among physical storage units, ensuring balanced erasing and writing operations, and optionally migrating data when significant differences in erasing counts are detected, thereby extending the service life of SSDs and improving system stability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is written into SSDs using conventional methods, then writing speed is improved, but some SSDs are subjected to excessive erasing and writing operations leading to premature damage

Engineering Contradiction:
Improvewriting speedVSAvoidSSD service life
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent changes the selection parameter for data writing from conventional methods (e.g., random selection, round-robin) to selection based on cumulative erasing count. The controller monitors and compares the cumulative erasing counts of different SSDs and dynamically selects the SSD with the lowest count for new data writing, thereby balancing wear across all SSDs in the array and extending overall system reliability

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements a feedback mechanism where the controller continuously monitors the cumulative erasing count of each SSD in the storage array. This feedback information is used to dynamically adjust the data writing distribution strategy, ensuring that SSDs with lower wear are selected for new writes, thus preventing any single SSD from being overused and extending the service life of the entire storage system

Inventive Principle:
Principle #23Feedback

2Device complexity

If no high availability solution is applied, then system complexity is reduced, but data loss risk increases due to premature SSD damage

Engineering Contradiction:
Improvesystem complexityVSAvoiddata loss risk
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent implements a self-service wear-balancing mechanism where the storage system automatically monitors its own SSD wear levels and redistributes data writing operations accordingly. The controller autonomously tracks cumulative erasing counts and redirects new writes to less-worn SSDs without requiring external intervention or complex high-availability infrastructure, thereby maintaining data reliability while minimizing system complexity

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent performs preliminary wear balancing by proactively monitoring cumulative erasing counts before SSD failure occurs. By predicting which SSDs will reach their wear limit first and preemptively redirecting writes away from them, the system prevents data loss before it happens, eliminating the need for reactive high-availability solutions and reducing overall system complexity

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11287984B2Method and device for writing stored data into storage medium based on flash memory
Publication Date: 2022.03.29 BEIJING OCEANBASE TECHNOLOGY CO LTD
  • US11287984B2 patent drawing
  • US11287984B2 patent drawing
  • US11287984B2 patent drawing

AI summary

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for writing data in one or more storage units. One of the methods includes obtaining an erasing count for each of multiple storage units, wherein the erasing count equals a total count of erasing operations that have been performed on all blocks of the storage unit. The method further includes identifying one or more of the storage units that satisfy one or more conditions associated with writing data, determining a storage unit that has the smallest erasing count among the identified storage units, and writing the data in the determined storage unit.