Direct-Mapped Flash Storage Rebalancing via OS-Level Address Translation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage systems face inefficiencies in data management and storage operations, particularly in handling flash storage systems where direct-mapped flash storage systems lack address translation by storage controllers, leading to redundant write operations and reduced reliability.
Innovation Solution
Implementing a direct-mapped flash storage system where the operating system initiates and controls processes across flash drives without address translation by storage controllers, utilizing non-volatile memory express (NVMe) for quick data access and erasure coding for redundancy, and employing zoned storage devices for efficient data allocation and wear leveling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional storage controllers are used with flash drives, then address translation is provided, but redundant write operations occur and reliability is reduced
Solution Approach 1:
The patent extracts the address translation function from the storage controller and relocates it to the operating system. The OS directly manages the mapping between virtual and physical addresses on flash drives, eliminating the controller's translation layer that caused redundant writes. This extraction resolves the contradiction by removing the source of redundancy while maintaining translation capabilities.
Solution Approach 2:
The patent introduces the operating system as an intermediary between the application layer and the flash storage. Instead of controllers handling address translation, the OS acts as the mediator that directly manages address mappings. This intermediary approach eliminates redundant write operations by providing a more efficient translation mechanism at the OS level rather than at the controller level.
2Productivity
If direct-mapped flash storage systems are implemented without address translation by storage controllers, then redundant write operations are minimized, but the system complexity increases due to OS-level address translation requirements
Solution Approach 1:
The patent applies universality by having the operating system perform multiple functions: it manages address translation, coordinates data writes, and handles storage optimization. By consolidating these functions in the OS rather than distributing them across controllers, the system achieves better write efficiency while the complexity is centralized in a single multi-functional component that all storage devices must interface with.
3Speed
If NVMe is used for quick data access, then data access speed is improved, but the need for optimized data allocation and wear leveling increases
Solution Approach 1:
The patent implements preliminary action by having the operating system proactively manage data allocation and wear leveling before performance degradation occurs. The OS monitors storage conditions and performs preventive optimization operations, such as redistributing data to balance wear across NVMe drives, ensuring sustained high performance without requiring complex real-time adjustments during data access operations.
Data Source
AI summary
Rebalancing in a fleet of storage systems using data science including generating, by the cloud-based rebalancing system, a plurality of workload migration scenarios to address a detected workload imbalance among a plurality of workloads in a fleet of storage systems; determining, by the cloud-based rebalancing system, a plurality of movement vectors for each workload migration scenario, wherein each of the plurality of movement vectors describes a consideration factor for migrating a workload of the plurality of workloads within the fleet of storage systems; and generating, by the cloud-based rebalancing system, at least one rebalancing proposal based on the plurality of movement vectors for each workload migration scenario.