Dynamic Immutability for Cloud Snapshot Data Retention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage of snapshot data in public cloud environments is vulnerable to unauthorized access, modification, or deletion, and marking data as immutable can prevent garbage collection, leading to increased storage costs and uncertainty about when snapshot data entities can be deleted due to complex dependencies and changes over time.
Innovation Solution
Implementing dynamic immutability for snapshot data entities by initially marking them as immutable for a limited period and periodically scanning to extend immutability only for entities not eligible for garbage collection, ensuring that immutability does not extend too far into the future, allowing for flexible retention and deletion based on retention periods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If snapshot data entities are marked as immutable to protect from unauthorized access and modification, then data security and integrity are improved, but garbage collection is prevented leading to increased storage costs
Solution Approach 1:
The patent applies dynamics by transitioning the immutability state from static (permanently immutable) to dynamic (temporarily immutable with time-bound retention periods). Snapshot data entities are marked immutable only for a specified retention period, after which they become eligible for garbage collection. This dynamic approach allows the system to balance security needs during the retention period with storage cost efficiency after expiration.
Solution Approach 2:
The patent implements periodic action through time-bound retention periods for immutable snapshot data. Instead of permanent immutability, the system applies immutability periodically for defined durations, allowing snapshot data to be protected during critical retention windows while automatically becoming eligible for deletion after expiration, thus enabling periodic garbage collection.
2Reliability
If snapshot data entities are marked as immutable to prevent deletion, then protection against ransomware attacks is improved, but uncertainty about deletion timing increases due to complex dependencies
Solution Approach 1:
The patent applies preliminary action by pre-defining retention periods for snapshot data entities before they become eligible for deletion. The system establishes immutability with a predetermined time boundary, allowing garbage collection to proceed systematically once the retention period expires, thereby eliminating uncertainty about deletion timing while maintaining protection during the retention window.
3Reliability
If immutability is extended far into the future for snapshot data entities, then data protection is improved, but flexibility for garbage collection and retention management is reduced
Solution Approach 1:
The patent applies parameter changes by modifying the immutability parameter from indefinite or extended future protection to a specific time-bound retention period. This parameter adjustment allows the system to maintain data protection during the defined retention window while enabling flexible garbage collection and retention management after expiration, thus balancing protection needs with operational flexibility.
Data Source
AI summary
A data management system may support techniques for immutable storage of snapshot data entities, which may each include data corresponding to one or more snapshots, in a cloud environment. The data management system may determine respective retention periods and respective immutability periods for the snapshot data entities. The data management system may extend the respective immutability period for a first snapshot data entity based on the respective retention period for the first snapshot data entity being greater than or equal to a threshold duration. Additionally or alternatively, the data management system may maintain (refrain from extending) the respective immutability period for a second snapshot data entity based at least in part on the respective retention period for the second snapshot data entity being less than the threshold duration.