Storage Controller Logical Physical Capacity Decoupling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage systems face inefficiencies due to the tight coupling of logical and physical capacity release, leading to suboptimal performance and increased resource consumption during capacity adjustments.
Innovation Solution
Decoupling and reordering of logical and physical capacity release in a storage system, allowing for independent management of logical and physical capacity, where logical capacity is made available immediately and physical capacity is released based on accumulated dereferencing operations, optimizing the release process to match the physical layout of storage devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If logical capacity release is tightly coupled with physical capacity release, then capacity management is simplified, but system performance deteriorates due to processing bottlenecks
Solution Approach 1:
The patent segments the capacity release process into two independent sub-processes: logical capacity release and physical capacity release. Logical capacity release handles address space management while physical capacity release handles actual storage block deallocation. This segmentation allows each sub-process to be optimized independently, resolving the contradiction between management simplicity and processing performance.
Solution Approach 2:
The patent introduces an intermediary mechanism (dereferencing operation accumulator) that buffers and batches physical capacity release operations. This intermediary decouples the immediate logical release from the deferred physical release, allowing logical operations to complete quickly while physical operations are processed in optimized batches, thereby maintaining both simplicity and performance.
2Reliability
If physical capacity is released immediately upon logical address release, then storage efficiency is improved, but system performance deteriorates due to excessive metadata updates
Solution Approach 1:
The patent performs preliminary accumulation of dereferencing operations before executing physical capacity release. Instead of immediately releasing physical blocks upon logical address release, the system accumulates multiple dereferencing operations and executes them in batches. This preliminary action reduces the frequency of metadata updates while ensuring storage efficiency is ultimately achieved.
Solution Approach 2:
The patent implements periodic execution of physical capacity release operations rather than continuous immediate release. By batching dereferencing operations and executing them periodically when thresholds are met, the system reduces metadata update overhead while maintaining storage efficiency, thereby resolving the contradiction between reliability and productivity.
3Device complexity
If dereferencing operations are executed in the same order as logical address release, then processing is simpler, but performance deteriorates due to non-optimized physical block access patterns
Solution Approach 1:
The patent inverts the traditional processing order by accumulating dereferencing operations and then reordering them based on physical block layout before execution. Instead of processing in the order logical addresses are released, the system reorders operations to match physical storage organization (e.g., sequential disk tracks or flash memory blocks). This inversion optimizes physical access patterns and significantly improves capacity release efficiency.
Data Source
AI summary
A storage system in one embodiment comprises a plurality of storage devices configured to store user data pages and metadata pages. Each of the user data pages has a logical address and a content-based signature derived from content of that data page, and each of the metadata pages characterizes a plurality of the user data pages and associates the content-based signatures of those user data pages with respective physical blocks in the storage devices. In conjunction with release of logical address space in the storage system, the released logical address space is made available to users in a first order based at least in part on released logical address, and multiple dereferencing operations are accumulated for respective ones of the physical blocks corresponding to the released logical address space. The accumulated dereferencing operations for the physical blocks are executed in a second order that differs from the first order.


