Dual FTL Memory Controller Bad Block Replacement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory controllers face challenges in efficiently managing reserved blocks across different memory areas, leading to reduced lifespan and performance due to inadequate bad block replacement strategies.

Innovation Solution

A memory controller with a dual flash translation layer (FTL) system that dynamically allocates and manages reserved blocks between memory areas, ensuring optimal block replacement and extending the lifespan of the memory device by requesting and allocating reserved blocks based on the number of available blocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single FTL manages all memory blocks, then the management structure is simple, but the lifespan and performance are reduced due to inadequate bad block replacement strategies

Engineering Contradiction:
ImproveFTL management structureVSAvoidmemory device lifespan
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent divides the FTL into two independent components: a first FTL for managing first memory blocks and a second FTL for managing second memory blocks. Each FTL independently tracks and manages reserved blocks for its respective memory area, enabling specialized bad block replacement strategies for each segment while maintaining overall system reliability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic reserved block allocation where the second FTL can allocate reserved blocks to the first FTL based on real-time bad block occurrence. The system dynamically adjusts the distribution of reserved blocks between the two FTLs, allowing flexible adaptation to varying bad block rates and extending overall memory device lifespan

Inventive Principle:
Principle #15Dynamics

2Ease of operation

If reserved blocks are not dynamically allocated between memory areas, then the management process is simple, but bad blocks cannot be effectively replaced leading to premature read-only mode

Engineering Contradiction:
Improvereserved block management processVSAvoidmemory device operational status
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent establishes a feedback mechanism where the first FTL monitors bad block occurrence and sends requests to the second FTL for additional reserved blocks when needed. The second FTL responds by allocating reserved blocks from its pool to the first FTL, creating a dynamic feedback loop that maintains adequate reserved block availability and prevents premature read-only mode

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent pre-allocates reserved blocks in the second memory area that can be dynamically transferred to the first memory area when bad blocks occur. This preliminary preparation of spare reserved blocks ensures that replacement can occur immediately without complex real-time management operations

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11360707B2Memory controller for replacing bad blocks with reserved blocks and operating method thereof
Publication Date: 2022.06.14 SK HYNIX INC
  • US11360707B2 patent drawing
  • US11360707B2 patent drawing
  • US11360707B2 patent drawing

AI summary

A memory controller includes a first flash translation layer (FTL) generating a physical address corresponding to a first type logical address received from a host on the basis of information about the first memory blocks, a second FTL generating a physical address corresponding to a second type logical address received from the host on the basis of information about the second memory blocks, and a memory control unit controlling the first memory area or the second memory area to perform an operation on the physical address corresponding to the first type logical address or the physical address corresponding to the second type logical address, wherein the first FTL provides the second FTL with block request information for requesting use of the second memory blocks, and generates the physical address corresponding to the first type logical address received from the host on the basis of block allocation information provided by the second FTL.