Memory Superblock Allocation for Defect-Aware Reclaim Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face challenges in effectively allocating superblocks of varying sizes due to the presence of defective blocks, leading to inefficiencies in data management and performance when handling mixed types of data.
Innovation Solution
A memory system that includes a controller managing superblocks with varying sizes by allocating write destinations based on Reclaim Units (RUs) designated by the host, allowing for efficient data placement and reuse of superblocks through Flexible Data Placement (FDP), ensuring data types are segregated and superblocks are utilized optimally.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If superblocks of varying sizes are used to accommodate defective blocks, then resource utilization is improved, but allocation complexity increases due to mismatch with fixed-size logical segments
Solution Approach 1:
The patent divides a superblock into multiple variable-size logical segments based on the number of defective blocks. When defective blocks are present, the superblock is segmented into a first logical segment for good blocks and a second logical segment for defective blocks, allowing flexible allocation that matches the actual usable capacity while maintaining manageable allocation units.
Solution Approach 2:
The patent implements dynamic segmentation where the number and size of logical segments within a superblock changes based on the detected number of defective blocks. The controller dynamically adjusts the segmentation scheme (e.g., 1 segment when no defects, 2 segments when defects exist) to optimize resource utilization while keeping allocation complexity adaptive rather than static.
2Ease of operation
If fixed-size logical segments are used for data allocation, then data management is simplified, but superblock resource utilization deteriorates due to size mismatch with varying superblock capacities
Solution Approach 1:
The patent applies local quality by making the size of logical segments dependent on their location and the defect configuration within the superblock. Different logical segments have different sizes tailored to the actual usable capacity of their corresponding physical blocks, ensuring optimal utilization while maintaining structured data management through defined segment boundaries.
3Productivity
If superblocks are allocated based on maximum size, then performance is improved through parallel operation, but waste increases when defective blocks reduce actual writable capacity
Solution Approach 1:
The patent changes the parameter of logical segment size based on the actual number of defective blocks in the superblock. By adjusting segment sizes to match the true writable capacity rather than using fixed maximum sizes, the system maintains high parallel operation performance through efficient allocation while reducing write amplification by avoiding allocation of space corresponding to defective blocks.
Data Source
AI summary
A memory system that can effectively use resources includes a nonvolatile memory that includes a plurality of physical blocks and a controller. The controller manages a plurality of superblocks that each include one or more physical blocks among the plurality of physical blocks, and notifies a host of a maximum size among data sizes writable in each of the superblocks, as a first size that is a size of a reclaim unit. Upon receipt of one or more write commands from the host, the one or more write commands designating a first reclaim unit having the first size, the controller writes a first data portion of the first reclaim unit into the first superblock to fill the entire first superblock and then a second data portion of the first reclaim unit into the second superblock.