Dynamic Data Preservation for HPB-LPB Storage Buffer Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semiconductor storage devices face performance degradation due to limited single-level cell (SLC) blocks, as QLC blocks are configured to function as SLC blocks, reducing available SLC blocks and degrading write and read performance when fill ratio exceeds optimal levels.
Innovation Solution
Implement a dynamic data preservation (DDP) threshold policy that adjusts based on fill ratio, identifying important data in high-performance buffer (HPB) blocks and migrating less important data to low-performance buffer (LPB) blocks during idle time, using a machine learning model to optimize read and write performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If QLC blocks are configured to function as SLC blocks to increase available high-performance blocks, then write performance is improved, but the number of available SLC blocks is reduced and performance degrades when fill ratio exceeds optimal levels
Solution Approach 1:
The patent implements dynamic data preservation by continuously monitoring the fill ratio of HPB blocks and dynamically adjusting the DDP threshold value. When the fill ratio exceeds the predefined preservation fill ratio, the system dynamically lowers the DDP threshold to migrate less important data to LPB blocks, thereby maintaining optimal HPB space utilization without fixed constraints
Solution Approach 2:
The system changes the parameter of DDP threshold value based on the fill ratio condition. The threshold is adjusted from a higher value when fill ratio is low to a lower value when fill ratio exceeds the preservation threshold, enabling adaptive control of data preservation behavior to match actual storage conditions
2Speed
If data is preserved in HPB blocks to maintain fast access, then read performance is improved, but write performance degrades when HPB blocks are full
Solution Approach 1:
The patent applies local quality by differentiating data into important and less important data based on access patterns and workload characteristics. Important data is preserved in HPB blocks for fast access, while less important data is migrated to LPB blocks, creating different storage qualities for different data types within the same system
Solution Approach 2:
The system performs preliminary data migration during idle time periods before write operations occur. By proactively migrating less important data from HPB to LPB blocks during idle periods, the system prepares HPB space in advance for upcoming write operations, preventing performance degradation
3Productivity
If DDP threshold value is lowered to migrate data during idle time, then HPB block space is freed for upcoming writes, but data preservation is reduced
Solution Approach 1:
The system implements feedback control by continuously monitoring the fill ratio of HPB blocks and using this information to adjust the DDP threshold value. The feedback loop ensures that data migration decisions are based on actual storage conditions, balancing space freeing with data preservation needs
Data Source
AI summary
An apparatus provides dynamic data preservation in a storage device, in which the apparatus includes control module configured to monitor a fill ratio of the storage device, compare the fill ratio with a predefined preservation fill ratio of the storage device to determine whether the fill ratio exceeds the predefined preservation fill ratio, dynamically configure a dynamic data preservation (DDP) threshold value of the storage device based on the fill ratio of the storage device when the fill ratio exceeds the predefined preservation fill ratio, and identify important data among data stored in high performance buffer (HPB) blocks of the storage device and migrate remaining data which are not identified as important data to low performance buffer (LPB) blocks of the storage device based on the configured DDP threshold value during idle time of the storage device.