Checkpoint Data Replication With Ordered Metadata Replay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies in data management and reliability, particularly in flash storage systems, due to redundant write operations and lack of direct control by the operating system, leading to increased wear and reduced reliability.
Innovation Solution
Implementing a direct-mapped flash storage system where the operating system directly addresses data blocks without translation by storage controllers, initiating and controlling processes such as data rewriting and erasure, and utilizing dual storage array controllers for failover and distributed management of erase blocks and metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If storage controllers translate and manage data blocks, then data management is centralized and simplified, but unnecessary write operations increase and reliability decreases
Solution Approach 1:
The patent extracts the translation function from the storage controller to the operating system. The OS directly addresses flash memory blocks without controller translation, eliminating the controller as an intermediary that causes redundant writes and reliability issues. This is achieved by having the OS manage erase blocks and metadata directly, bypassing the traditional controller architecture.
Solution Approach 2:
The operating system performs self-service by directly managing flash memory operations including translation, erasure, and data placement. The OS maintains its own translation layer and erase block management, eliminating dependence on storage controllers for these critical functions, thereby reducing unnecessary writes and improving reliability.
2Ease of operation
If storage controllers manage data rewriting and erasure, then control is centralized, but write operations increase and flash lifespan decreases
Solution Approach 1:
The patent extracts the data rewriting and erasure management functions from the storage controller to the operating system. The OS directly controls when and how data is rewritten and erased, eliminating controller-induced redundant operations that waste flash cycle life. The OS maintains direct control over erase block allocation and data placement.
Solution Approach 2:
The operating system performs self-service flash management by directly handling rewriting and erasure operations. The OS monitors flash wear, manages erase block lifecycles, and controls data placement without controller intervention, thereby minimizing unnecessary write operations and extending flash storage lifespan.
3Device complexity
If single storage controller is used, then system is simpler, but failover capability and reliability are reduced
Solution Approach 1:
The patent segments the storage control function across multiple independent storage controllers. Each controller manages specific flash drives and erase blocks, creating a distributed control architecture. This segmentation enables failover capability where surviving controllers can take over failed controller's managed resources, improving reliability without excessive complexity.
4Measurement precision
If operating system directly addresses data blocks without translation, then control precision and reliability improve, but device complexity increases
Solution Approach 1:
The patent introduces an OS-level translation layer as an intermediary between the application and flash memory. This translation layer resides in the operating system rather than in hardware controllers, enabling precise direct addressing of flash blocks while maintaining the necessary address translation functionality. The OS mediator manages erase block mapping and data placement with high precision.
Data Source
AI summary
Continuous data protection, including sending, to a target data repository from a source data repository, metadata describing one or more updates to one or more datasets stored within the source data repository; generating, based on the metadata describing the one or more updates to the one or more datasets, an ordered log of metadata describing an ordered application of the one or more updates to the one or more datasets; and generating, on the source data repository and based on the ordered log of metadata, the one or more datasets in accordance with the one or more updates corresponding to a specified point in time.