Penalty Index Workload Allocation in Networked Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Efficiently allocating resources in networked storage environments to optimize workload placement is challenging due to the complexity of managing dissimilar but related components such as storage nodes, storage pools, and inter-connects, which requires a systematic approach to determine the optimal subset of resources for processing workloads.
Innovation Solution
A management console determines a penalty index for subsets of resources based on supply parameters, threshold values, and criticality factors to automatically assign workloads to the subset with the least penalty index, thereby optimizing resource allocation and workload placement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual resource allocation methods are used in networked storage systems, then flexibility and control are maintained, but the complexity of managing dissimilar components (storage nodes, storage pools, inter-connects) increases and optimal workload placement becomes difficult to achieve
Solution Approach 1:
The system implements self-service through automated penalty index calculation and workload placement decisions. The management console automatically evaluates resource subsets, calculates penalty indices based on supply parameters and criticality factors, and assigns workloads without manual intervention, allowing the storage system to manage itself optimally
Solution Approach 2:
The invention changes the parameter representation by introducing a unified penalty index parameter that aggregates multiple resource characteristics (supply parameters, threshold values, criticality factors) into a single comparable metric, enabling systematic optimization across dissimilar components through parameter transformation
2Measurement precision
If automated penalty index calculation is implemented for all resource subsets, then optimal workload placement is achieved, but the computational effort and processing time increase
Solution Approach 1:
The system segments the resource evaluation process by dividing resources into hierarchical subsets (storage nodes, storage pools, inter-connects) and calculating penalty indices for each subset independently. This segmentation allows parallel processing and reduces the computational complexity of evaluating all resources simultaneously
Data Source
AI summary
In a networked storage system, a request for placing a workload for is received by a management console that determines a plurality of subset of resources for responding to the workload. Each subset of resources includes at least two different but related resource types that can be used for processing the workload. A penalty index for each subset of resources is then determined. The penalty index of each resource is based on a supply parameter that indicates a maximum capability of the resource, a threshold value that indicates a capability level beyond which no load is added to the resource and a criticality factor that indicates an importance of each resource to the networked storage system. The workload is then assigned to a subset of resources with the least penalty index.