Storage Data Relocation by Access Mode
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In hierarchized disk systems, data block relocation across SSD, SAS, and SATA disks leads to degraded sequential access performance and data fragmentation due to the lack of information on access modes, causing inefficiencies in storage hierarchization control.
Innovation Solution
A storage system that identifies the access mode of data blocks as sequential or random, using a program to determine and manage data block relocation based on access frequency and mode, ensuring that sequential access data blocks are maintained in a sequential order to prevent fragmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data blocks are relocated based on access frequency in hierarchized disk systems, then storage performance is optimized, but sequential access performance degrades and data fragmentation occurs
Solution Approach 1:
The patent segments data blocks into two categories: sequential access data blocks and random access data blocks. This segmentation allows different relocation strategies to be applied to each type, preventing the degradation of sequential access performance while still optimizing storage performance through targeted relocation of random access blocks.
Solution Approach 2:
The patent applies local quality by treating sequential access data blocks differently from random access data blocks in the relocation process. Specifically, sequential access blocks are excluded from relocation or given priority maintenance of their sequential order, while random access blocks are freely relocated based on access frequency. This localized differentiation resolves the contradiction between optimization and performance maintenance.
2Productivity
If data blocks are relocated across different disk types, then storage utilization is optimized, but data fragmentation increases
Solution Approach 1:
The patent segments the data block population into sequential and random access categories, applying different relocation rules to each segment. Random access data blocks are relocated across disk types based on access frequency to optimize storage utilization, while sequential access data blocks are protected from relocation or maintained in sequential order, thereby preventing data fragmentation.
Solution Approach 2:
The patent applies local quality by implementing location-specific relocation policies: random access data blocks are freely relocated to optimize storage utilization across different disk types, while sequential access data blocks are given special treatment to maintain their sequential integrity. This localized approach allows storage optimization without causing fragmentation.
3Ease of operation
If relocation is performed without considering access mode, then relocation simplicity is maintained, but sequential access performance degrades
Solution Approach 1:
The patent applies preliminary action by determining the access mode of data blocks before executing relocation operations. The system identifies and categorizes data blocks as sequential or random access type in advance, then applies appropriate relocation strategies. This preliminary classification adds minimal complexity while effectively preventing sequential access performance degradation.
Solution Approach 2:
The patent implements self-service by enabling the storage system to automatically identify and classify data blocks based on their access patterns. The system monitors access modes and autonomously determines which data blocks should be relocated and which should be protected, eliminating the need for manual intervention while maintaining sequential access performance.
Data Source
AI summary
A non-transitory computer-readable recording medium has stored therein a program for causing a computer to execute a process. The process includes identifying a data block from among a plurality of data blocks in a first storage for relocation to a second storage, determining an access mode of the identified data block, the access mode including sequential access or random access, and relocating the identified data block to the second storage based on the determined access mode.