Workload Classification for Storage Resource Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud resource management systems fail to differentiate between storage types when executing operations, leading to inefficient use of high-quality storage resources and increased costs, as they assume that more high-quality resources always result in better performance without considering task priority and historical utilization trends.
Innovation Solution
A method that prioritizes workloads by classifying them as high or low priority based on activity levels and migrates low-priority workloads from high-end to low-end storage resources, optimizing resource allocation and reducing competition for high-end storage resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If more high-quality storage resources are allocated to all workloads, then overall system performance improves, but infrastructure costs increase significantly
Solution Approach 1:
The patent applies local quality by differentiating storage resource allocation based on workload characteristics. High-priority workloads receive high-quality storage resources while low-priority workloads are allocated to low-end storage resources. This selective allocation optimizes system performance for critical tasks while reducing overall infrastructure costs, avoiding the need to upgrade all storage resources uniformly.
2Speed
If high-end storage resources are used for all workloads, then data access speed improves, but resource utilization efficiency deteriorates
Solution Approach 1:
The system implements local quality by matching storage resource quality to workload priority levels. High-priority workloads that require fast data access are allocated to high-end storage resources, while low-priority workloads are directed to low-end storage resources. This differentiation maintains optimal data access speed for critical operations while significantly improving overall resource utilization efficiency.
Solution Approach 2:
The patent applies partial action by providing high-quality storage resources only to the extent necessary for high-priority workloads, rather than universally. The system selectively applies performance optimization where it is most needed, avoiding excessive resource allocation to low-priority tasks that do not require high-speed access.
3Device complexity
If storage resources are not differentiated by priority, then system simplicity is maintained, but performance consistency across different workload types deteriorates
Solution Approach 1:
The patent applies segmentation by dividing workloads into distinct priority categories (high-priority and low-priority) and allocating appropriate storage resources to each segment. This classification mechanism, implemented through the activity analyzer and workload classifier components, ensures performance consistency across different workload types while maintaining manageable system complexity through automated decision-making.
Data Source
AI summary
A method includes storing a plurality of workloads in a first disk resource associated with a high end disk classification. The method further includes determining a corresponding activity level for each of the plurality of workloads. The method also includes classifying each of the plurality of workloads into a first set indicative of high-priority workloads and a second set indicative of low-priority workloads based on whether the corresponding activity level is greater than a threshold activity level. The method further includes determining whether a second disk resource associated with a low end disk classification can accommodate storage of a first particular workload in the second set based on an available storage capacity of the second disk resource. The method additionally includes migrating the first particular workload from the first disk resource to the second disk resource.