Storage Deduplication Control via Write Frequency Indexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The combination of deduplication and hierarchization techniques in storage systems leads to inefficient use of storage space and increased garbage collection processing, as data blocks with low access frequencies are written to slower devices and overwritten, causing wasteful storage usage and reduced performance.
Innovation Solution
An information processing apparatus that calculates an index value based on the write count of data blocks and determines whether to perform deduplication, securing a new area for frequently written data blocks and avoiding deduplication on blocks written at the same address, thereby optimizing storage usage and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If deduplication technique is applied to storage, then storage cost is reduced, but storage performance deteriorates when data blocks are frequently overwritten
Solution Approach 1:
The patent dynamically adjusts the deduplication policy based on the write frequency of data blocks. By monitoring how often data is written to specific addresses, the system switches between performing deduplication (for low-frequency writes) and holding deduplication (for high-frequency writes), thereby optimizing both storage efficiency and performance adaptively
Solution Approach 2:
The patent changes the parameter of deduplication application by introducing an index value that represents write frequency. When the index value exceeds a threshold, the system changes its behavior from performing deduplication to holding it, thus optimizing performance for frequently written data while maintaining storage efficiency for other data
2Productivity
If hierarchization technique is applied after deduplication, then storage performance is improved, but storage space is wastefully used
Solution Approach 1:
The patent performs preliminary action by calculating the index value (write frequency) before applying hierarchization. This allows the system to identify frequently written data blocks in advance and prevent their duplication, thereby avoiding wasted storage space while still enabling performance optimization through hierarchization
3Quantity of substance
If deduplication is performed on frequently written data blocks, then storage efficiency is improved, but garbage collection processing increases
Solution Approach 1:
The patent implements feedback by continuously monitoring the write frequency of data blocks and using this information to adjust the deduplication policy. The index value serves as a feedback mechanism that informs the system when to perform or hold deduplication, thereby reducing unnecessary garbage collection operations on frequently written data
Data Source
AI summary
An information processing apparatus includes a memory and a processor coupled to the memory. The processor is configured to calculate, when writing a first data block at a first address on a storage, a first index value corresponding to a number of times of writing data at the first address. The processor is configured to determine, on basis of the first index value, whether to perform or hold deduplication on the first data block. The processor is configured to switch between performing and holding the deduplication on the first data block depending on a result of the determination.