Storage Controller Selective Deduplication Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing storage systems face challenges in reducing data storage costs and maintaining high access performance due to varying compression ratios and inefficient deduplication processes, which can lead to increased storage costs and reduced performance.
Innovation Solution
A storage system that dynamically manages logical pages by performing deduplication on pages with low write frequency and reversing deduplication on pages with low deduplication ratios, using a combination of deduplication and Thin Provisioning techniques to optimize storage space and access performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If deduplication process is applied to all partitions, then storage area consumption is reduced, but access performance deteriorates due to processing overhead
Solution Approach 1:
The patent applies deduplication selectively to specific partitions based on their characteristics (write frequency, deduplication ratio) rather than uniformly to all partitions. The storage controller identifies candidate partitions with low write frequency and applies deduplication only to those, while maintaining high-performance access for frequently written partitions.
Solution Approach 2:
The patent dynamically adjusts the deduplication strategy based on partition characteristics and performance metrics. The storage controller monitors write frequency and deduplication ratios, and adaptively determines which partitions should undergo deduplication, allowing the system to optimize between storage efficiency and access performance in real-time.
2Quantity of substance
If compression is applied to reduce data size, then storage cost is reduced, but access performance deteriorates due to decompression overhead
Solution Approach 1:
The patent applies compression selectively to partitions with low write frequency where the overhead is less impactful, while avoiding compression on frequently accessed partitions. This localized approach ensures that compression benefits are realized without significantly degrading overall system performance.
3Quantity of substance
If deduplication is applied to partitions with high write frequency, then storage efficiency is improved, but access performance deteriorates due to frequent processing overhead
Solution Approach 1:
The patent specifically identifies and excludes partitions with high write frequency from deduplication processing. The storage controller monitors write frequency metrics and directs deduplication operations only to candidate partitions with low write frequency, thereby preserving high access performance while still achieving storage efficiency gains on appropriate partitions.
4Speed
If selective deduplication is applied based on write frequency, then access performance is maintained, but storage area consumption is not sufficiently reduced
Solution Approach 1:
The patent employs dynamic monitoring and adjustment of deduplication strategies. The storage controller continuously evaluates partition characteristics including write frequency and deduplication ratios, and adaptively determines which partitions should undergo deduplication. This dynamic approach allows the system to maximize storage efficiency while maintaining access performance by making informed, data-driven decisions about where to apply deduplication.
Solution Approach 2:
The patent incorporates feedback mechanisms where the storage controller monitors deduplication ratios and write frequency metrics, and uses this information to adjust deduplication strategies. When deduplication ratios are low, the system can adjust its behavior to avoid unnecessary deduplication overhead, while when ratios are high, it aggressively applies deduplication to maximize storage efficiency.
Data Source
AI summary
A storage system according to an aspect of the present invention includes one or more storage devices for storing write data to which a write request from a host computer is directed, and a storage controller that provides one or more volumes to the host computer. Further, the storage system manages the time when a write request is last received from the host computer for each partition within the volume. Then, the storage controller performs a deduplication process upon detecting the partition not receiving a write request for a predetermined time or more from the time when the write request is last received.