Dynamic Data Relocation Engine for Storage Tiering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data relocation systems in information handling systems relocate data at fixed times, which does not account for varying data access patterns, leading to inefficiencies and potential disruptions in data access for global user bases, especially in datacenters.
Innovation Solution
A data relocation engine that monitors data access frequencies and relocates data from one storage device to another based on predicted access patterns, moving data to higher or lower speed storage devices when its access frequency is about to exceed or fall below a threshold, thereby optimizing data placement and minimizing disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If data relocation is performed at fixed time intervals, then storage system operation is simplified, but data access efficiency deteriorates due to ignoring varying data access patterns
Solution Approach 1:
The patent implements dynamic data relocation by continuously monitoring data access frequencies and automatically relocating data between storage devices of different speeds based on real-time access patterns. This replaces the static fixed-time relocation approach with a dynamic system that adapts to changing data access requirements, thereby improving data access efficiency while maintaining operational simplicity through automated decision-making.
2Device complexity
If all data is relocated at a fixed time, then storage management is simplified, but data access interruptions increase for global user bases
Solution Approach 1:
The patent employs preliminary action by predicting future data access frequencies based on historical patterns and proactively relocating data before access interruptions occur. The system monitors access patterns and initiates relocation operations in advance when it detects that access frequency will exceed thresholds, thereby preventing data access interruptions rather than reacting to them after they occur.
Solution Approach 2:
The patent implements feedback mechanisms by continuously monitoring data access frequencies and using this information to dynamically adjust relocation decisions. The system collects real-time access data, compares it against predefined thresholds, and automatically triggers relocation operations when necessary, creating a closed-loop control system that maintains data access continuity while simplifying storage management.
3Productivity
If large amounts of data are relocated simultaneously at fixed times, then storage tiering is achieved, but internal I/O load peaks increase affecting user access
Solution Approach 1:
The patent applies segmentation by dividing the data relocation process into small, incremental operations based on individual data access patterns. Instead of relocating large volumes of data simultaneously at fixed intervals, the system segments relocation tasks into smaller units triggered by specific access frequency thresholds, thereby achieving storage tiering while distributing I/O load over time and avoiding peak loads that would affect user access.
Solution Approach 2:
The patent implements periodic action through continuous monitoring of data access frequencies with automatic trigger-based relocation. Rather than executing bulk relocation at fixed periodic intervals, the system performs periodic access pattern analysis and triggers relocation only when access frequency thresholds are exceeded, creating a responsive periodic mechanism that maintains storage tiering efficiency while avoiding concentrated I/O load peaks.
Data Source
AI summary
A data relocation system includes a storage controller device coupled to first storage device(s) that operate at first data access speeds, and second storage device(s) that operate at second data access speeds that are higher than the first data access speeds. During a first time period, the storage controller device monitors first data that is stored in the first storage device(s) to identify a first data access frequency for the first data, and determines that the first data access frequency for the first data during the first time period indicates that a second data access frequency for the first data will exceed a data access frequency threshold during a second time period that occurs immediately following the first time period. In response and during the second time period, the storage controller device relocates the first data from the first storage device(s) to the second storage device(s).