Cloud Data Object Locking via Periodic Garbage Collection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud storage systems are vulnerable to malicious attacks, particularly when credentials are compromised, leading to potential data deletion, and existing solutions for locking data objects are either inefficient or can be manually removed by attackers.
Innovation Solution
Implementing a cloud-based backup system with a compliance mode that uses a garbage collector to automatically and uniformly lock data objects for a predetermined time frame, ensuring locks cannot be removed by attackers, and extending or deleting them only during scheduled garbage collection runs, thereby preventing unnecessary long-term locking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data objects are locked to protect them from malicious attacks, then data security is improved, but storage costs increase due to long-term locking
Solution Approach 1:
The system implements periodic garbage collection runs that temporarily lock data objects only during the collection process, rather than maintaining continuous locks. This periodic, temporary locking provides security during vulnerable operations while avoiding the continuous storage costs of long-term locks.
Solution Approach 2:
The lock duration is made dynamic rather than static - objects are locked only for the duration needed during garbage collection operations, with lock timing automatically adjusted based on when collection is needed, providing adaptive security that minimizes unnecessary storage costs.
2Reliability
If locks are applied to data objects to prevent deletion, then data protection is improved, but system complexity increases due to manual management requirements
Solution Approach 1:
The garbage collection system automatically manages locking and unlocking of data objects without requiring manual intervention. The system self-regulates the locking process based on collection needs, eliminating complex manual lock management while maintaining data protection.
Solution Approach 2:
The system uses feedback from the garbage collection process to automatically control lock states - when collection is needed, locks are applied; when collection is complete, locks are removed. This automated feedback loop simplifies management by tying lock states directly to system needs rather than requiring external control.
3Reliability
If locks are extended for long periods to ensure security, then data safety is improved, but storage efficiency decreases due to unnecessary long-term locking
Solution Approach 1:
Instead of continuous long-term locking, the system applies locks periodically only when garbage collection is performed. This ensures data safety during collection operations while maintaining storage efficiency by unlocking objects during normal operations when they can be safely accessed and deleted if needed.
Solution Approach 2:
The system performs garbage collection at scheduled intervals before data might be needed, proactively identifying and removing unnecessary data while it is locked temporarily. This preliminary action approach ensures safety during the collection process while avoiding prolonged locks that would reduce storage efficiency.
Data Source
AI summary
A lock is applied automatically without any user involvement, to each of one or more data objects stored by the data processing system. Each of the one or more locks prevent deletion, by any user, of the one or more data objects within a predetermined time frame. A cloud garbage collector marks the one or more data objects for a) deletion, or b) for lock extension, the garbage collector scheduled to run periodically at a period that is less than the predetermined time frame. The deletion or the lock extension is executed for the one or more data objects, based on marks by the garbage collector. Other embodiments are described and claimed.