Pattern Repository Storage for Backup Data Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current disk-based data backup systems face performance issues due to poor access latency, which can lead to poor performance in restoring backup data, especially when versioning is used without considering the access latency of storage devices.
Innovation Solution
A method and apparatus that utilize a pattern repository to store and reference data patterns, balancing access time and storage efficiency by storing multiple copies of patterns based on access time, allowing for efficient compression and maintaining sequential stream performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If versioning is used to compress data by storing differences between versions, then storage capacity is reduced, but access latency increases and restore performance deteriorates
Solution Approach 1:
The patent segments data into fixed-size blocks and organizes them in a hierarchical structure with hot, warm, and cold storage tiers. This segmentation allows frequently accessed data blocks to be stored in fast storage while less frequently accessed blocks are moved to slower storage, resolving the contradiction between compression and access latency by spatially separating data based on access patterns.
Solution Approach 2:
The system performs preliminary actions by pre-calculating and storing metadata about data blocks including access frequency and timing information. This preliminary metadata preparation enables the system to make rapid decisions about data placement and retrieval without real-time analysis, reducing access latency while maintaining compression efficiency.
2Loss of time
If multiple copies of data patterns are stored to reduce access time, then storage capacity increases, but storage efficiency decreases
Solution Approach 1:
The patent applies local quality by creating multiple copies of data patterns only in specific locations (hot storage tier) where they are frequently needed, rather than uniformly distributing copies throughout the system. This localized replication reduces access time for critical data while minimizing overall storage capacity consumption by avoiding unnecessary copies elsewhere.
Solution Approach 2:
The system dynamically changes parameters such as replication factor and data placement location based on observed access patterns. When a data pattern is accessed frequently, the system increases its replication factor and places copies in faster storage; when access frequency decreases, it reduces replication and moves data to slower storage, optimizing the balance between access time and storage capacity.
3Quantity of substance
If data is compressed by storing only differences between versions, then storage efficiency improves, but restore performance deteriorates due to poor access latency
Solution Approach 1:
The patent introduces an intermediary component (the data management system with metadata database) that mediates between the compressed stored data and the restore operation. This intermediary maintains metadata that maps logical data blocks to their physical locations and version information, enabling rapid identification and retrieval of needed data blocks without requiring full decompression and comparison operations, thus improving restore performance while maintaining compression efficiency.
Data Source
AI summary
A method and an apparatus to store data patterns are presented. In one embodiment, the method includes searching a pattern repository to find prior copies of a pattern and to reference one of the prior copies, or insert a new copy, based on the access time of the prior copy and the effect on the sequential stream performance.