Storage Controller Data Allocation by Retention Strength
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage devices face challenges in extending their lifespan and performance due to high reclaim amounts, which are influenced by variations in retention strengths of physical units within non-volatile memory, leading to inefficient data retention and frequent reclaim operations.
Innovation Solution
A storage device and controller system that determines the hotness of data and allocates it to physical units based on retention strengths, writing hot data to units with lower retention strengths to delay reclaim operations and reduce the total reclaim amount, thereby improving retention time and device performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is written to physical units without considering retention strength variations, then the storage device structure remains simple, but the reclaim amount increases and lifespan decreases
Solution Approach 1:
The patent applies local quality by differentiating the treatment of different physical units based on their retention strengths. The storage controller classifies physical units into multiple groups according to their retention characteristics and assigns different allocation strategies to each group. This allows the system to optimize data placement for each local region (physical unit group) rather than treating all units uniformly, thereby reducing reclaim operations and extending lifespan without requiring complete system redesign
Solution Approach 2:
The patent implements preliminary action by pre-characterizing the retention strengths of physical units during manufacturing or initialization and storing this information in the storage controller. This advance preparation enables the controller to make informed allocation decisions when data needs to be written, avoiding the need for complex real-time analysis and reducing reclaim operations before they occur
2Reliability
If physical units with high retention strength are used for all data, then data retention is maximized, but frequent reclaim operations still occur due to process variations
Solution Approach 1:
The patent differentiates data allocation based on the specific retention characteristics of each physical unit group. By matching data with appropriate retention requirements to corresponding physical unit groups, the system achieves optimal retention for each data type while minimizing unnecessary reclaim operations across the entire storage device
Solution Approach 2:
The patent introduces dynamic allocation strategies where the storage controller adapts its data placement decisions based on the specific retention strength characteristics of available physical units. This dynamic approach allows the system to optimize retention performance while reducing reclaim frequency by matching data hotness levels with appropriate physical unit groups
3Loss of time
If hot data is written to physical units with lowest retention strength, then reclaim operations are delayed, but data may be lost if retention strength is insufficient
Solution Approach 1:
The patent pre-characterizes and stores retention strength information for all physical units in the storage controller before data allocation occurs. This preliminary characterization enables the controller to make informed decisions about which physical units to assign to hot data, ensuring that retention strength requirements are met while maximizing delay of reclaim operations
Solution Approach 2:
The patent implements feedback mechanisms where the storage controller continuously monitors data retention status and adjusts allocation decisions accordingly. By comparing actual retention performance against expected values and observing reclaim operation timing, the system can refine its allocation strategies to balance delay of reclaim operations with assurance of data retention reliability
Data Source
AI summary
Provided are a storage device, a storage controller, and an operating method of the storage controller. The storage device according to the inventive concept includes a non-volatile memory and a storage controller. The non-volatile memory includes a plurality of memory blocks, each memory block includes physical units having different retention strengths due to process variations, and the retention strengths respectively correspond to times that physical units retain data before respective reclaim operations for the physical units. The storage controller receives a write request and data from a host, selects a first physical unit based on hotness of data and retention strengths, and controls the non-volatile memory to write data to the first physical unit.