Predictive Data Object Retrieval via Hierarchical Storage Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage management solutions experience delays in data object movement from slower to faster storage devices, which can hinder process execution, especially for critical processes, due to the manual and cumbersome nature of existing prefetching methods.
Innovation Solution
Utilizing historical execution information to predict data requirements for a process and proactively move data from slower storage devices to faster ones before it is requested, implementing this within a hierarchical storage management system to stage data until process completion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If data objects are moved from slower storage devices to faster storage devices in response to process requests, then storage cost is reduced by using slower storage devices, but process execution is delayed due to data movement time
Solution Approach 1:
The system performs preliminary actions by analyzing historical execution information to predict future data requirements and proactively moving data objects from slower to faster storage devices before the process actually requests them. This prefetching mechanism eliminates execution delays while maintaining cost efficiency by only pre-moving predicted data.
Solution Approach 2:
The system uses historical execution information as feedback to continuously improve predictions of data requirements. By analyzing past process execution patterns, the system refines its ability to predict which data objects will be needed, enabling more accurate prefetching decisions that balance storage cost and execution speed.
2Loss of time
If customized functionality is implemented to retrieve specific data objects before process execution, then process execution delay is reduced, but system complexity and manual intervention requirements increase
Solution Approach 1:
The system performs self-service by automatically analyzing historical execution information, predicting data requirements, and moving data objects without manual administrator intervention. The storage management system autonomously identifies which data to prefetch and executes the transfer, eliminating the need for customized manual configurations while achieving fast data availability.
Solution Approach 2:
The system performs preliminary analysis of historical execution patterns to predict future data needs before processes run. This automated preliminary action replaces manual customization with intelligent forecasting, reducing both execution delays and system complexity.
Data Source
AI summary
A solution for managing data in which historical execution information for a process is used to determine a set of data requirements for the process. The set of data requirements is then used to move data for the process from a slower storage device (e.g., a magnetic tape) to a faster storage device (e.g., a hard drive) before the process requests the data object (e.g., prefetch). In this manner, the prefetched data can be provided when requested by the process without delaying its execution while the data is moved from the slower storage device to the faster storage device.