Hierarchical Storage Management for Memory Lifespan and Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage device management systems fail to efficiently utilize memory hierarchies by not effectively moving data between different types of storage devices based on access frequency, leading to suboptimal performance and memory degradation.
Innovation Solution
A storage device management apparatus and method that uses a memory controller to manage data across RAM, SLC NAND flash, and MLC NAND flash by prioritizing data storage and retrieval based on access frequency, moving data from RAM to SLC NAND flash and then to MLC NAND flash as needed, and performing garbage collection to maintain optimal storage conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is stored in high-performance RAM, then access speed is improved, but storage capacity is limited and cost increases
Solution Approach 1:
The storage system is segmented into multiple hierarchical levels (RAM, SLC NAND flash, MLC NAND flash) with different performance and capacity characteristics. Data is divided and stored across these segments based on access frequency, allowing the system to achieve both high speed for frequently accessed data and large capacity for less frequently accessed data.
Solution Approach 2:
The patent introduces a new dimension of storage hierarchy by adding intermediate storage layers (SLC and MLC NAND flash) between RAM and traditional secondary storage. This creates a multi-dimensional storage architecture that resolves the contradiction by providing multiple pathways for data storage with different performance-capacity trade-offs.
2Productivity
If data is moved between storage devices based on access frequency, then memory efficiency is improved, but device complexity increases
Solution Approach 1:
The storage management system operates autonomously by automatically monitoring data access patterns and triggering migrations without user intervention. The controller dynamically decides when and where to move data based on predefined policies, making the system self-managing and reducing operational complexity despite the sophisticated hierarchy.
Solution Approach 2:
The storage architecture is designed to be dynamic, with data automatically migrating between storage layers based on real-time access frequency. The system adapts its configuration continuously, adjusting which data resides in which storage medium based on changing access patterns, thereby optimizing efficiency without requiring complex manual management.
3Productivity
If frequently accessed data is kept in RAM, then access performance is improved, but memory degradation accelerates
Solution Approach 1:
The patent changes the physical state and location parameter of data by migrating it between different storage media with different durability characteristics. Less frequently accessed data is moved to more durable but slower storage media (NAND flash), reducing the wear on RAM while maintaining access performance for frequently used data.
Solution Approach 2:
The patent introduces intermediate storage layers (SLC and MLC NAND flash) that act as mediators between RAM and traditional secondary storage. These intermediaries absorb some of the access load and protect RAM from excessive write operations, thereby extending RAM lifespan while maintaining system performance.
Data Source
AI summary
According to one embodiment, a storage device management device is connected to a random access memory and a first storage device. When the random access memory includes a free region sufficient to store write data, the write data is stored onto the random access memory. Data on the random access memory selected in the descending order of elapsed time from the last access is sequentially copied onto the first storage device, and a region in the random access memory which has stored the copied data is released. When stored on the random access memory, the read data is read from the random access memory to the processor. When stored on the first storage device, the read data is copied onto the random access memory and read from the random access memory to the processor.