Unified Datacenter Storage Model for Fabric Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing complex datacenter environments with interconnected devices and varied workloads is challenging due to inconsistencies in hardware configurations and differing expectations of underlying fabric behavior, making it difficult to create and maintain accurate models for deployment and management.
Innovation Solution
A method that involves creating infrastructure and application models to understand how specific services are deployed on uniquely identified hardware, facilitating the management of applications, datacenters, and the datacenter fabric by modeling hardware and virtual components, and their interconnections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If administrators manage complex datacenter environments with interconnected devices and varied workloads, then the datacenter functionality and service capability are enhanced, but the complexity of configuration and management increases significantly
Solution Approach 1:
The patent segments the datacenter management system into distinct modular components: fabric discovery module, workload discovery module, modeling module, and deployment module. Each module handles specific aspects of discovery and management independently, reducing overall system complexity while maintaining comprehensive functionality.
Solution Approach 2:
The patent introduces an intermediary modeling layer that sits between the physical fabric/workloads and the management interface. This model layer abstracts complex interconnections and relationships into standardized representations, making the system easier to manage while preserving full functionality.
2Ease of operation
If consistent hardware configurations are implemented across workloads, then management becomes easier, but the flexibility to meet different workload expectations and requirements is reduced
Solution Approach 1:
The patent applies local quality by allowing different workload models to have different properties and requirements. Each workload can be configured with specific fabric dependencies, resource requirements, and deployment constraints tailored to its needs, while the overall system maintains standardized management processes.
Solution Approach 2:
The patent enables parameter changes by allowing workload models to specify different fabric configuration parameters, resource allocation parameters, and deployment parameters. This allows consistent management approaches while adapting to diverse workload requirements through parameter variation rather than structural change.
3Measurement precision
If detailed models of fabric and workloads are created, then deployment accuracy and troubleshooting capability are improved, but the time and resources required to create and maintain models increase
Solution Approach 1:
The patent implements preliminary action by automatically discovering fabric and workload configurations and pre-building models before deployment is needed. The system proactively maintains model accuracy through continuous discovery updates, eliminating the need for manual model creation and reduction maintenance time.
Solution Approach 2:
The patent enables self-service by implementing automated discovery mechanisms that continuously update fabric and workload models without administrator intervention. The system self-maintains model accuracy by automatically detecting changes and updating representations, significantly reducing maintenance time and resources.
Data Source
AI summary
Modeling an application deployed in a distributed system. The method includes accessing an infrastructure model of a distributed system. The infrastructure model includes a model of specific physical hardware including unique identifiers for each piece of hardware and an identification of interconnections of the physical hardware. The method further includes accessing an application model for an application. The application model defines the components that make up the application and how the components are to be deployed. The method further includes deploying the application in the distributed system by deploying elements of the application on hardware modeled in the infrastructure model. The method further includes using the infrastructure model and the application model deployment creating a deployment model defining how the application is deployed on the physical hardware.