Storage Controller Super Block Allocation Strategy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage devices face inefficiencies when handling sequential and random write requests, leading to underutilization of super blocks and deterioration of over-provisioning lifespan.

Innovation Solution

A storage device with a controller that dynamically allocates and deallocates super blocks based on the counts of sequential and random write requests, ensuring optimal usage and extending the lifespan by preventing prolonged underutilization of super blocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If separate super blocks are allocated for sequential and random write requests, then operational performance is improved when both types are received simultaneously, but availability of over-provisioning deteriorates when only one type is continuously received

Engineering Contradiction:
Improveoperational performanceVSAvoidavailability of over-provisioning
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements dynamic super block allocation where the controller adjusts super block assignment based on real-time write request patterns. When sequential writes dominate, sequential super blocks are utilized; when random writes dominate, random super blocks are utilized. This dynamic adaptation allows the system to maintain high performance for the dominant write type while preserving over-provisioning availability by deallocating unused super blocks back to the pool.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the allocation state parameter of super blocks based on monitored write request characteristics. By detecting whether sequential or random write requests are predominant, the controller modifies the allocation status of super blocks accordingly, transitioning between allocated and deallocated states to optimize both performance and over-provisioning availability.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If super blocks are continuously allocated for specific write request types, then performance for that write type is optimized, but super blocks remain underutilized when the other write type is received

Engineering Contradiction:
Improvewrite operation performanceVSAvoidsuper block underutilization
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The controller implements a feedback mechanism that continuously monitors the types and patterns of write requests received from the host. Based on this feedback, the controller dynamically adjusts super block allocation decisions, dealingloca ting super blocks when they are not needed for the current dominant write type, thereby eliminating underutilization while maintaining performance optimization.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Super blocks are designed to be multi-functional, capable of serving both sequential and random write operations. The controller assigns appropriate super blocks based on the current write request type, allowing the same pool of super blocks to fulfill different functional requirements dynamically, thus preventing underutilization while maintaining specialized performance characteristics.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250123764A1Storage device for managing super block and method of operating the same
Publication Date: 2025.04.17 SK HYNIX INC
  • US20250123764A1 patent drawing
  • US20250123764A1 patent drawing
  • US20250123764A1 patent drawing

AI summary

Provided herein is a semiconductor memory device having improved lifespan. The storage device may include a memory device including a plurality of super blocks, and a controller configured to control the memory device in response to each request among a plurality of requests from a host, wherein the controller is further configured to allocate a first super block to a first write request of a first type, and allocate a second super block different from the first super block to respond to a second write request of a second type, control the memory device to store in the first super block write data corresponding to the first write request, and when the second write request is received from the host before reallocation of the second super block after deallocation, control the memory device to store in the first super block write data corresponding to the second write request.