Policy-Based Storage Scheduling for Heterogeneous Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing and provisioning heterogeneous storage systems is complex, as existing technologies typically support only homogeneous systems, making it difficult to control and manage multiple data services across different storage environments efficiently.
Innovation Solution
A policy-based scheduling model is introduced that maps data storage resources within a heterogeneous environment to a data storage host, creating I/O paths and enabling unified management through a REST API and CLI, allowing for the conversion of multiple storage arrays into a single logical array, scalable to millions of storage volumes and file shares.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a policy-based scheduling model is implemented to manage heterogeneous storage resources, then the ease of operation and unified management is improved, but the device complexity increases due to the need for modeling objects, transport zones, and I/O paths
Solution Approach 1:
The patent introduces a policy-based scheduling model as an intermediary layer between storage resources and storage hosts. This model includes modeled instances of objects that represent physical storage resources, transport zones for mapping resources to hosts, and I/O path configurations. The intermediary layer abstracts the complexity of heterogeneous storage environments, enabling unified management through standardized policies while hiding the underlying complexity from users.
2Adaptability or versatility
If multiple storage arrays are federated into a single logical array, then the adaptability and scalability are improved, but the device complexity increases due to the need for resource provisioning and mapping mechanisms
Solution Approach 1:
The patent creates a universal interface and standardized model that can manage diverse storage arrays from different vendors and environments. The policy-based scheduling model uses universal object instances, transport zones, and I/O path definitions that work across heterogeneous storage systems. This multi-functional approach allows the same management framework to handle various storage array types, protocols, and configurations, enabling scalability without proportionally increasing complexity.
3Productivity
If automated provisioning is implemented through policy-based scheduling, then the productivity is improved, but the device complexity increases due to the need for sophisticated scheduling algorithms and object modeling
Solution Approach 1:
The patent implements self-service automation where the policy-based scheduling model automatically provisions storage resources based on predefined policies and object models. The system autonomously creates transport zones, maps storage resources to hosts, configures I/O paths, and manages resource allocation without manual intervention. This self-service capability dramatically improves provisioning productivity while the complexity is contained within the automated system rather than requiring complex manual procedures.
Data Source
AI summary
A computer-executable method, system, and computer program product comprising providing a policy based scheduling model comprising modeled instances of objects representing resources, wherein the resources are provisioned within a heterogeneous storage environment, wherein the modeled instances of objects includes a transport zone for mapping a data storage resource within the heterogeneous storage environment to a data storage host, wherein each of the modeled instances of objects includes one or more ports to create I/O paths between the data storage resource and the data storage host, and using the policy based scheduling model to manage the provisioning of the data storage resource in the heterogeneous storage environment.