Storage System Data Movement Based on Access Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The long latency of nonvolatile memories in storage devices, such as NAND flash memory, degrades performance and necessitates techniques to improve storage device performance.
Innovation Solution
A storage system with multiple memory regions, including a host system and storage sets with first, second, and third memory regions, where data from the third memory region with long latency is moved to either the first or second memory region based on data access features, such as immediate use, frequent access, or importance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is stored in nonvolatile memory (third memory region), then data retention and reliability are improved, but access latency increases and performance degrades
Solution Approach 1:
The storage system is segmented into multiple memory regions with different characteristics: volatile memory regions (first and second memory regions) for fast access and nonvolatile memory region (third memory region) for data retention. This segmentation allows data to be stored in the nonvolatile memory for reliability while enabling quick access by keeping frequently used data in volatile memory regions.
Solution Approach 2:
The controller acts as an intermediary between the host system and the multiple memory regions. It manages data movement between volatile and nonvolatile memory regions, automatically transferring data based on access patterns. This intermediary function resolves the contradiction by handling the complexity of data placement transparently to the host system.
2Productivity
If data is moved frequently between memory regions based on access patterns, then performance is improved, but system complexity increases
Solution Approach 1:
The system implements self-service through automatic data movement based on access patterns. The controller monitors data access behavior and autonomously decides when to move data between memory regions without requiring complex external management. This reduces the operational complexity while maintaining performance benefits.
Solution Approach 2:
The system performs preliminary actions by pre-loading data into volatile memory regions before it is actually needed. Based on predicted access patterns, the controller proactively moves data from nonvolatile to volatile memory, reducing actual access latency without requiring complex real-time decision-making during data access operations.
Data Source
AI summary
Disclosed are a storage system and an operation method therefor. The storage system includes: a host system; and a plurality of storage sets configured to interface with the host system. At least one of the plurality of storage sets includes: a first memory region; a second memory region; and a third memory region, and the at least one of the plurality of storage sets is configured to move data stored in the third memory region to a selected memory region among the first memory region and the second memory region based on a data access feature.