Hierarchical Storage Management Global Policy Redistribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data storage systems face challenges in efficiently managing large data transfers within limited 'backup windows' due to increasing data volumes, requiring complex networked systems that often lead to stagnant or waiting jobs, and existing management components primarily focus on data collection rather than proactive operation optimization.
Innovation Solution
A hierarchical data storage system with a global manager that monitors data storage operations, redistributes jobs and resources based on load reports, and implements global policies and filters to optimize data transfer processes, allowing for dynamic adjustment of storage operations and resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data storage systems transfer and store large amounts of data during limited backup windows, then data storage capacity is improved, but system complexity increases and operations become less efficient
Solution Approach 1:
The system segments storage operations into multiple hierarchical levels (global management layer, cell management layer, and storage device layer). Each level handles specific tasks independently, allowing the system to manage large data volumes without proportionally increasing overall system complexity. Local cells can autonomously manage their operations while the global layer coordinates resource allocation.
Solution Approach 2:
The patent introduces a temporal dimension to storage operations by implementing load report mechanisms that track system state over time. This allows the system to make intelligent scheduling decisions based on historical patterns, distributing operations across different time periods within backup windows to optimize throughput without requiring proportional increases in system complexity.
2Adaptability or versatility
If data storage operations occur at different places and times in networked systems, then system versatility is improved, but job stagnation and waiting increase
Solution Approach 1:
The system implements comprehensive feedback mechanisms through load reports that continuously monitor storage operation status across all cells. This feedback enables the global management layer to dynamically reallocate jobs between cells based on current load conditions, preventing job stagnation while maintaining the versatility of distributed operations across different locations and times.
Solution Approach 2:
The patent makes the storage system dynamic by allowing job redistribution and resource allocation to change in real-time based on system conditions. Cells can transition between idle and active states, and job priorities can be adjusted dynamically, enabling the system to adapt to varying workloads and minimize waiting times across the distributed network.
3Measurement precision
If management components collect data from various storage components, then measurement capability is improved, but operational optimization is insufficient
Solution Approach 1:
The system enables self-service operation optimization where the global management layer automatically processes load report data and makes scheduling decisions without requiring external intervention. The management components not only collect data but also autonomously analyze it and adjust operations accordingly, transforming passive data collection into active operational optimization that directly improves productivity.
Data Source
AI summary
A system and method for setting global actions in a data storage system is described. In some examples, the system determines a policy based on information from the system, and implements that policy to the system. In some examples, the system adds or modifies global filters based on information from the system.