QLC Memory Compaction Strategy Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory sub-systems using quad-level cell (QLC) memory face inefficiencies due to the need for initial data storage in single-level cell (SLC) memory as cache, which occupies significant space and degrades quality of service, especially in enterprise and zoned namespace applications.
Innovation Solution
Implementing a memory sub-system that manages QLC compaction strategies based on device characteristics, such as QLC level batch, SLC level, and partial SLC level compaction strategies, to optimize data storage and retrieval by selecting the appropriate compaction strategy based on storage space and usage, thereby reducing cache requirements and improving performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is initially stored in SLC memory as cache before programming to QLC memory, then data storage reliability is improved, but available host space is reduced and quality of service is degraded
Solution Approach 1:
The patent implements dynamic compaction strategies that adaptively adjust the amount of SLC cache allocated based on device characteristics, workload types, and usage patterns. The system can switch between different compaction approaches (e.g., aggressive compaction for sequential workloads, conservative compaction for random workloads) to optimize the balance between reliability and available space in real-time.
Solution Approach 2:
The system changes key parameters including compaction thresholds, cache allocation sizes, and programming schedules based on device characteristics such as wear level, error rates, and performance metrics. By dynamically adjusting these parameters, the system maintains data reliability while maximizing available host space for different application scenarios.
2Manufacturing precision
If SLC memory is used as cache for QLC programming, then programming precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts the complex compaction management logic from the hardware layer and implements it in the firmware/software layer. This allows the physical device structure to remain relatively simple while achieving sophisticated compaction control through software algorithms that monitor device state and adjust programming strategies accordingly.
Solution Approach 2:
The system introduces an intermediary compaction management layer that sits between the host and the memory device. This intermediary handles the complex tasks of monitoring device characteristics, selecting appropriate compaction strategies, and coordinating SLC/QLC programming operations, thereby shielding the host from complexity while maintaining programming precision.
3Area of stationary object
If aggressive QLC compaction is performed, then available host space is increased, but quality of service management becomes difficult
Solution Approach 1:
The patent implements comprehensive feedback mechanisms that continuously monitor quality of service metrics such as read latency, write throughput, and error rates. The compaction management system uses this feedback to adjust compaction aggressiveness dynamically - backing off when QOS degradation is detected and intensifying compaction when metrics are healthy, thereby maintaining both high host space utilization and quality of service.
Data Source
AI summary
One of a plurality of compaction strategies to be performed on the memory device based on at least one characteristic of a memory device is identified. Each of the plurality of compaction strategies is to program host data from at least one single-level cell (SLC) of the memory device to at least one quad-level cell (QLC) of the memory device. One or more host data from a host system is received. A compaction operation on the one or more host data using the one of the plurality of compaction strategies is performed.