Storage Systems with Independent Security Threat Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage systems face inefficiencies in data management and security, particularly in handling write operations and maintaining data integrity across multiple flash drives, leading to potential data loss and system instability.
Innovation Solution
The implementation of a direct-mapped flash storage system where the operating system initiates and controls data allocation and garbage collection across multiple flash drives, reducing reliance on lower-level storage controllers and minimizing redundant write operations, while utilizing non-volatile RAM as a buffer for improved latency and power resilience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional storage controllers manage data allocation and garbage collection, then storage management is automated, but write operations become redundant and efficiency decreases
Solution Approach 1:
The patent extracts the data allocation and garbage collection functions from the storage controller and transfers them to the operating system. This eliminates redundant write operations through controller intervention while maintaining automated management at a higher software layer, directly resolving the contradiction between write efficiency and controller complexity.
Solution Approach 2:
The operating system directly manages flash drive allocations and performs garbage collection without requiring storage controller intervention. This self-service approach eliminates unnecessary write amplification while maintaining automated storage management, improving write operation efficiency without proportionally increasing OS complexity.
2Reliability
If data is directly written to flash drives without buffering, then write speed is fast, but data integrity is compromised during power failures
Solution Approach 1:
The patent introduces non-volatile RAM as an intermediary buffer between the host interface and flash drives. Data is first written to the NVRAM buffer, ensuring integrity protection during power failures, then asynchronously transferred to flash drives. This mediator resolves the contradiction by providing both data protection and reduced write latency through batching.
Solution Approach 2:
The system performs preliminary data validation and buffering in NVRAM before committing data to flash drives. This preliminary action ensures data integrity is established early in the write process, while the buffered approach allows multiple writes to be consolidated, reducing overall write latency.
3Reliability
If synchronous replication is implemented across storage systems, then data consistency is improved, but write operations become slower due to coordination overhead
Solution Approach 1:
The patent implements preliminary data buffering in NVRAM at both primary and secondary storage systems before replication. This allows write operations to be acknowledged once data is safely buffered locally, with replication occurring asynchronously in the background. Data consistency is maintained through coordinated buffer management, while write speed is improved by eliminating synchronous replication wait times.
Solution Approach 2:
The system uses feedback mechanisms where the operating system monitors buffer status and replication state across distributed storage systems. This feedback enables intelligent decisions about when to acknowledge writes and when to perform replication, maintaining data consistency while optimizing write performance based on real-time system conditions.
Data Source
AI summary
A system may include a plurality of storage systems each configured to operate in accordance with a synchronous replication configuration in which a dataset is synchronously replicated to each of the plurality of storage systems such that each of the plurality of storage systems stores and continually updates a different copy of the dataset. Each of the storage systems may be configured to independently monitor, while operating in accordance with the synchronous replication configuration, for security threats against one of the copies of the dataset; and perform, when the monitoring indicates that the one of the copies of the dataset is possibly being targeted by a security threat, a remedial action with respect to the security threat.