Multi-Tier Storage Data Block Relocation Based on Row Age
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage management solutions in multi-tier storage systems often relocate data blocks containing both newer and older rows to lower tiers, which can degrade performance when newer rows need to be accessed, as they are moved together with older rows.
Innovation Solution
Identify data blocks with a mix of older and newer rows and prevent them from being relocated to lower tiers, and monitor blocks after relocation; if a new row is added, move them back to a higher tier, using time-based values or other criteria to determine tier placement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If data blocks containing both newer and older rows are relocated to lower tiers, then storage costs are reduced, but performance degrades when newer rows need to be accessed
Solution Approach 1:
The patent segments a data file into multiple data blocks, where each block contains a portion of the table data. This allows selective relocation of individual blocks to different storage tiers based on their access patterns, rather than moving entire files or tables. Blocks containing primarily older rows can be moved to lower-cost tiers while blocks with newer rows remain in higher-performance tiers.
Solution Approach 2:
The patent applies different storage tier placements to different parts (blocks) of the same data file based on local characteristics. Each block is evaluated independently for its access frequency and row age distribution, allowing optimized placement decisions at the block level. This enables newer rows to be kept in high-performance tiers while older rows are moved to lower-cost tiers within the same file.
2Loss of energy
If entire data files are moved to lower tiers, then storage costs are reduced, but performance degrades due to mixed older and newer rows in the same blocks
Solution Approach 1:
The patent divides data files into smaller blocks that can be independently managed and relocated. This segmentation allows the system to move only the necessary blocks to lower tiers rather than entire files, maintaining performance consistency for blocks containing newer rows while still achieving cost savings for blocks with older rows.
Solution Approach 2:
The patent applies partial action by relocating only specific blocks within a data file rather than the entire file. This selective approach moves just the blocks that contain primarily older rows to lower tiers, while leaving blocks with newer rows in higher-performance tiers, thus achieving cost optimization without sacrificing performance consistency.
3Speed
If blocks are monitored after relocation and moved back when new rows are added, then performance is maintained, but system complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where the system monitors data blocks after relocation to lower tiers. When new rows are added to a block that has been moved, the system detects this change and automatically relocates the block back to a higher-performance tier. This feedback loop ensures performance is maintained while using automated rules to minimize management complexity.
Solution Approach 2:
The monitoring and relocation system operates autonomously based on predefined rules. When new rows are detected in a relocated block, the system automatically triggers the relocation back to a higher tier without requiring manual intervention. This self-service approach maintains performance while reducing the operational burden on storage administrators.
Data Source
AI summary
In general, a block of data in a data file is stored in a multi-tier storage system. The block of data includes multiple rows and multiple entry values per row, including values for a particular entry. The values of the particular entry in the data block can be used to determine whether to move the data block to a different tier of a multi-tier storage system. The block of data can then either be relocated in a different tier or kept in the current tier.