Content Addressable Storage Deduplication via Time-Partitioned Data Release
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The deletion process in existing storage systems with deduplication is complicated by distributed storage configurations and failure tolerance requirements, leading to increased space consumption and performance issues, as traditional garbage collection methods are not effective in these environments.
Innovation Solution
A storage system that utilizes a data storage controlling unit to manage storage target data and address data based on content, allowing deduplication to occur concurrently with deletion, and a data release controlling unit that releases storage regions not pointed to by other data, using a time-series approach to optimize storage utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional garbage collection methods are used in distributed storage systems with deduplication, then space reclamation can be performed, but the deletion process becomes complicated and increases space consumption
Solution Approach 1:
The patent segments the storage data into different time zones (current time zone and past time zones) to enable selective deletion. By dividing data temporally, the system can delete only unnecessary past data while preserving current data, simplifying the deletion process and improving space utilization without increasing overall complexity.
Solution Approach 2:
The patent performs preliminary organization of data into time zones before deletion occurs. By pre-organizing data structurally with clear temporal boundaries, the deletion process becomes simpler and more efficient, avoiding the complexity of managing mixed temporal data during deletion operations.
2Productivity
If deduplication is performed on backup streams on-the-fly, then writing of duplicated blocks can be avoided and performance improves, but deletion becomes significantly complicated
Solution Approach 1:
The patent segments data into time zones to enable independent management of current and past data. This segmentation allows the system to maintain on-the-fly deduplication performance for current data while simplifying deletion by targeting only past data that doesn't need to be preserved, thus resolving the contradiction between performance and deletion complexity.
Solution Approach 2:
The time zone structure acts as an intermediary layer between the deduplication mechanism and the deletion process. It mediates between the need for continuous deduplication performance and the complexity of deletion operations, allowing both functions to operate efficiently without interfering with each other.
3Quantity of substance
If space reclamation is performed immediately, then storage utilization improves, but system performance and reliability are compromised
Solution Approach 1:
The patent performs preliminary organization of data into time zones, which enables deferred space reclamation. By pre-organizing data structurally, the system can safely delay immediate reclamation while maintaining data accessibility, and then perform reclamation in a controlled manner that preserves reliability.
Solution Approach 2:
The patent introduces dynamic time zone management that allows the system to adaptively balance between space utilization and data accessibility. The time zone structure enables flexible control over when data can be deleted, allowing the system to maintain reliability during active periods and perform space reclamation when safe to do so.
Data Source
AI summary
In a content address storage system, storage target data or address data is stored in a storage device with respect to each time zone divided in a time-series manner, and a storage region in the storage device storing a piece of data, which is not pointed to by other address data, of the storage target data or the address data stored in the storage device in a past time zone before a current time zone, is released.