Storage System Control Reducing Write Wear via Partial Logging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAID systems incur additional overhead and latency due to the need for redundant data logging during write operations, which increases write wear and reduces throughput, as they require writing data blocks twice for crash consistency and fault tolerance.
Innovation Solution
A method is proposed where only a portion of the dataset is logged for data recovery, rather than the entire dataset, allowing for reduced logging overhead and improved storage system performance by synergizing fault tolerance and crash consistency mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire dataset is logged into the storage system for crash consistency and fault tolerance, then data recovery capability is improved, but write wear increases and throughput decreases
Solution Approach 1:
The patent segments the dataset into multiple data blocks and selectively logs only a portion of them (first number of data blocks) rather than the entire dataset. This segmentation allows the system to maintain sufficient data for recovery while reducing the logging overhead that would otherwise reduce throughput.
Solution Approach 2:
The patent applies partial action by logging only a subset of data blocks (first number) rather than all data blocks (second number, where first number < second number). This partial logging approach provides adequate fault tolerance and crash consistency while avoiding the excessive write operations that would increase write wear and reduce throughput.
2Reliability
If redundant data is maintained among storage nodes for fault tolerance, then fault tolerance capability is improved, but additional overhead is caused
Solution Approach 1:
The patent merges the fault tolerance mechanism with the crash consistency mechanism by using the same logged data blocks to serve dual purposes: maintaining fault tolerance through redundant data storage and ensuring crash consistency through the logging process. This consolidation eliminates the need for separate redundant data maintenance operations that would add overhead.
Solution Approach 2:
The logged data blocks serve multiple functions simultaneously: they provide crash consistency information, enable fault tolerance through redundancy, and reduce write wear by avoiding unnecessary duplicate writes. This multi-functionality reduces the overall overhead of maintaining the storage system.
3Reliability
If data blocks are written twice for crash consistency, then crash consistency is improved, but write wear increases
Solution Approach 1:
The patent extracts the essential crash consistency requirement from the full data writing process. Instead of writing all data blocks twice, it identifies and logs only the critical data blocks (first number of data blocks) that are necessary for crash consistency, thereby reducing unnecessary write operations and write wear.
Solution Approach 2:
The patent changes the parameter of data blocks to be logged from the total number of data blocks to a subset (first number < second number). This parameter change maintains the crash consistency guarantee while reducing the volume of data written to the log, thereby decreasing write wear.
Data Source
AI summary
A method for controlling a redundant storage system is proposed. A write request to a redundant storage system is received (310). A dataset that is to be written into the redundant storage system by the write request is determined (320). A portion of the dataset is logged into the redundant storage system for data recovery in case that a fault occurs in the redundant storage system (330). Thus, only a portion of the dataset is logged and the amount of the logged data may be reduced compared with the conventional redundant storage system. Further, the redundant storage system may be recovered in response to a fault occurring in the redundant storage system.