Storage Array Data Organization Reducing Cleaning Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face inefficiencies in managing snapshots, leading to time-consuming data writes and reads due to the mixing of current and superseded data in metadata maps, which complicates space recovery and compression processes.
Innovation Solution
The method involves selectively organizing memory page groups by generating metadata lookup structures to correlate disk addresses with logical block addresses, creating owner groups for current and subsequent states of storage volumes, and assigning reference access until new data writes supersede older data, allowing for efficient data management and compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is compressed into containers in a given map, then storage economy is improved, but cleaning overhead increases because containers may include both currently active and superseded data
Solution Approach 1:
The patent segments data into distinct categories: currently active data and superseded data. By organizing these segments separately within the map structure, the system can compress both types of data efficiently while identifying and cleaning only the superseded portions, thereby reducing cleaning overhead while maintaining storage economy.
2Ease of operation
If maps search from newest to oldest to find current data, then data retrieval is possible, but read operations become time-consuming because any map may contain both current and superseded data
Solution Approach 1:
The patent applies preliminary action by organizing data in maps with clear indicators of current versus superseded status before read operations occur. This pre-organization allows the system to quickly identify and access current data without performing exhaustive searches through all maps, significantly reducing read operation time while maintaining retrieval capability.
3Quantity of substance
If older snapshots are deleted to recover space, then storage space is recovered, but the process is complicated by the mix of current and superseded data in each map
Solution Approach 1:
The patent extracts superseded data from containers and identifies it as separate from currently active data. This extraction enables the system to delete only the superseded portions when recovering space from old snapshots, simplifying the space recovery process by eliminating the need to navigate through mixed current and superseded data in each map.
Data Source
AI summary
A system and method reduces cleaning overhead in a storage array by establishing owner groups for selectively organizing memory page groups. The method generates a metadata lookup structure to correlate a disk address of the real-time data generated to a logical block address on the select storage volume. Further, the method establishes a current state of the select storage volume as a first owner group that receives reference access to the logical block address associated with the memory page group. The method selectively generates a memory snapshot of the current state of the select storage volume to form a subsequent owner group. Additionally, the method enables the first owner group and the subsequent owner group to reference the logical block address that is associated with the memory page group until a real-time data write provides a new logical block address that is associated with a new memory page group.