Storage Class Selection Using Predicted Data Longevity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies in managing data storage and retrieval due to redundant write operations and lack of centralized control over storage processes, leading to reduced reliability and increased wear on storage devices.
Innovation Solution
Implementing a direct-mapped flash storage system where higher-level processes, such as the operating system, manage data allocation and storage operations directly without involving lower-level storage controllers, and utilizing non-volatile RAM for buffering data before writing to persistent storage to reduce latency and improve reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If storage controllers manage data allocation and storage operations, then storage operations are automated, but device complexity increases and redundant writes occur
Solution Approach 1:
The patent extracts the storage management functions from the storage controller and places them in the host system. The host directly manages data allocation, address translation, and storage operations, removing the intelligent controller layer that causes complexity and redundant writes. This is achieved by implementing a controller-less storage architecture where the host OS handles all storage management tasks.
Solution Approach 2:
The patent introduces a buffer memory (DRAM) as an intermediary between the host and flash storage devices. This buffer mediates data transfers, allowing the host to directly manage storage operations while the buffer handles temporary data storage and address translation, eliminating the need for complex storage controllers.
2Reliability
If storage controllers perform write operations, then data is persisted, but device wear increases due to redundant writes
Solution Approach 1:
The patent implements preliminary actions by having the host system determine data longevity and set appropriate write policies before data is written to storage. The host predicts which data will be modified soon and prevents writing such data to flash storage, performing the prediction and policy decision in advance to avoid redundant writes and extend device lifespan.
Solution Approach 2:
The patent implements feedback mechanisms where the system monitors storage device wear and data access patterns, then adjusts write operations accordingly. The host receives feedback about data longevity predictions and modifies write behavior to prevent redundant writes, creating a closed-loop system that optimizes both data persistence and device lifespan.
3Quantity of substance
If multiple storage devices are managed independently, then storage capacity is increased, but storage efficiency decreases due to lack of centralized control
Solution Approach 1:
The patent merges multiple storage devices into a unified storage pool managed directly by the host. Instead of each device being independently controlled, the host consolidates management of all flash storage devices, enabling efficient data placement, wear leveling, and load distribution across the entire pool, thereby improving storage efficiency while maintaining increased capacity.
Data Source
AI summary
Cost considerate placement of data within a pool of storage resources, including: receiving one or more data objects for storage; selecting, based at least upon a storage policy and upon one or more characteristics of storage data, one or more storage classes from among a plurality of storage classes of one or more data storage services, wherein the storage policy specifies parameters for one or more of: storage costs, storage operation response time, data resiliency, or service level agreement specifications; and storing the one or more data objects to the selected one or more storage classes of the one or more data storage services.