Storage Task Scheduling by Use Levels and Priority
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage system management technologies face challenges in efficiently scheduling resources among various task types executed on different virtual storage layers, leading to inefficient allocation and utilization of physical storage resources.
Innovation Solution
A method that determines use levels and priority levels of task types based on current and historical resource use statuses across multiple storage layers, allowing for the selection and sorting of tasks for execution to optimize resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple task types share physical storage resources across different virtual storage layers, then resource utilization increases, but resource allocation efficiency deteriorates due to lack of differentiated scheduling
Solution Approach 1:
The patent segments task scheduling by introducing task types and storage layer associations. Tasks are divided into different categories (e.g., sequential tasks, random tasks) and associated with specific storage layers (e.g., hot data layer, cold data layer). This segmentation enables differentiated resource allocation while maintaining overall resource sharing, thus improving allocation efficiency without reducing utilization.
Solution Approach 2:
The patent implements dynamic task scheduling by continuously monitoring resource usage patterns and adjusting task execution schedules in real-time. The scheduling mechanism adapts to changing workloads by dynamically selecting which tasks to execute and when, optimizing resource allocation efficiency while maintaining high utilization across shared physical storage resources.
2Productivity
If tasks are executed on corresponding storage layers according to task types, then task execution efficiency improves, but system complexity increases due to multiple scheduling dimensions
Solution Approach 1:
The patent creates a universal scheduling framework that handles multiple task types and storage layer associations through a single integrated mechanism. The scheduling system performs multiple functions simultaneously: task classification, storage layer mapping, resource allocation, and execution scheduling. This multi-functionality approach improves task execution efficiency while avoiding the complexity of multiple separate scheduling systems.
Solution Approach 2:
The patent simplifies the scheduling system by changing key parameters from complex multi-dimensional scheduling to a streamlined model based on task type and storage layer association. By defining clear parameter relationships (task type → storage layer → resource allocation), the system achieves high task execution efficiency with reduced scheduling complexity compared to general-purpose scheduling approaches.
3Productivity
If use levels and priority levels are determined based on current and historical resource use statuses, then resource allocation optimality improves, but computational overhead increases
Solution Approach 1:
The patent performs preliminary classification of tasks into task types and pre-establishes associations with storage layers before execution. By preparing task metadata and determining scheduling parameters in advance, the system reduces real-time computational overhead while maintaining optimal resource allocation based on current and historical resource usage patterns.
Solution Approach 2:
The patent replaces complex real-time computational analysis with a rule-based scheduling mechanism that uses pre-determined task types and storage layer associations. Instead of performing heavy computational optimization at runtime, the system uses predetermined classification rules and historical data patterns to achieve near-optimal resource allocation with significantly reduced computational overhead.
Data Source
AI summary
Techniques are directed to storage system management within a storage system. Such storage system management involves task scheduling. Along these lines, a group of tasks for execution may be selected based on task type use levels and priority levels. Accordingly, there may be more reasonable allocation of resources thus increasing the utilization of IO bandwidth.