Storage Controller Data Relocation via Temporal Proximity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data storage systems face performance bottlenecks due to the temporal proximity of accesses to data residing on the same storage device or accessed via a common controller, which existing horizontal and vertical distribution methods fail to adequately address.
Innovation Solution
A data storage system that records associations between data accessed within a defined service window and migrates data to different storage devices to distribute access via different hardware channels, reducing latency by separating temporally proximate access requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is distributed horizontally among multiple storage devices, then I/O throughput is increased, but access latency increases due to temporal proximity of accesses via common hardware channels
Solution Approach 1:
The system performs preliminary analysis of access request patterns to identify temporally proximate accesses to different data. By detecting these patterns in advance and proactively relocating data before the bottleneck occurs, the system prevents latency issues rather than reacting to them. The storage controller monitors access requests and identifies when multiple accesses targeting different data are scheduled within the same service window, then initiates data relocation to different storage devices ahead of time.
Solution Approach 2:
The system adds a temporal dimension to the traditional horizontal data distribution approach. Instead of only considering spatial distribution across storage devices, the patent introduces time-based analysis by examining service windows and temporal proximity of accesses. This temporal dimension allows the system to intelligently determine when and where to relocate data, optimizing both throughput and latency by considering the timing of access requests alongside the physical location of data.
2Loss of time
If data is distributed vertically among multiple tiers, then access latency is reduced, but I/O throughput decreases due to access bottlenecks at common controllers
Solution Approach 1:
The system segments the access paths by relocating data to different storage devices that have separate hardware channels. This segmentation divides the previously shared access path into multiple independent paths, allowing simultaneous accesses to proceed without contending for the same hardware resources. By segmenting the access paths at the storage device level rather than just at the controller level, the system eliminates the bottleneck while maintaining low latency.
Solution Approach 2:
The storage controller acts as an intermediary that monitors access patterns and makes intelligent relocation decisions. It detects when multiple access requests targeting different data are scheduled within the same service window and initiates relocation to different storage devices. This intermediary function allows the system to dynamically optimize the distribution of data based on actual access patterns, balancing throughput and latency requirements.
3Loss of time
If data relocation is performed frequently to optimize access patterns, then access latency is reduced, but system complexity increases due to migration management
Solution Approach 1:
The system employs periodic action by analyzing access requests in discrete service windows and making relocation decisions based on patterns detected within each window. Rather than continuously monitoring and relocating, the system periodically evaluates access patterns and takes action when beneficial conditions are met. This periodic approach reduces the complexity of continuous monitoring while still achieving latency optimization through targeted relocations.
Solution Approach 2:
The system uses feedback from monitored access patterns to drive relocation decisions. By detecting when multiple accesses to different data are scheduled within the same service window, the system receives feedback about potential bottlenecks and responds by relocating data to eliminate the conflict. This feedback-driven approach ensures relocations are performed only when necessary, reducing unnecessary complexity while maintaining performance optimization.
Data Source
AI summary
A data storage system includes first and second storage devices accessed via first and second hardware channels, respectively. A storage controller receives a first access request requesting access to first data and a second access request requesting access to second data, where both the first data and the second data are accessed via the first hardware channel. In response to detecting receipt of the first and second access requests within a same service window defining a range of temporal proximity, the storage controller records an association of identifiers of the first data and second data requested by the first and second access requests, respectively. The storage controller thereafter migrates the second data accessed by the second access request from the first storage device to the second storage device based on the recorded association, such that the first and second data are available for access via different hardware channels.