Data Center Resource Allocation via Multi-Tier Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional resource allocation in data centers is often inefficient, as it focuses on individual tiers and may result in insufficient path bandwidth or unreliable switches, and existing optimization frameworks face limitations such as brittleness, infeasibility, inaccuracies, and complexity in representing system behavior and optimizing resource allocation.
Innovation Solution
A computer-implemented method and system using a consolidated virtualization workbench that receives application information and data center resource features to create possible configurations, correlating white-box and black-box models to predict relationships of cost, risk, and performance, and performing multiple-dimensional analysis to select optimal configurations for data center resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional individual-tier resource allocation is used, then storage capacity and performance requirements can be satisfied, but path bandwidth insufficiency and unreliable switch connections occur
Solution Approach 1:
The patent combines server, storage, and network resource allocation into a unified multi-tier optimization framework. The system simultaneously considers and optimizes resources across all three tiers together rather than independently, ensuring that storage selection accounts for network path bandwidth and switch reliability, thereby resolving the contradiction between allocation reliability and coordination complexity.
2Productivity
If comprehensive multi-tier optimization is implemented, then resource allocation quality improves, but system complexity and computational burden increase
Solution Approach 1:
The patent segments the complex multi-tier optimization problem into hierarchical layers: first optimizing storage resources based on capacity and performance requirements, then optimizing network resources based on path bandwidth and switch reliability, and finally optimizing server resources. This segmentation reduces computational complexity while maintaining comprehensive optimization across all tiers.
Solution Approach 2:
The system performs preliminary optimization of storage resources before proceeding to network and server optimization. By pre-determining storage allocations that satisfy capacity and performance requirements, the system reduces the search space for subsequent network and server optimization, thereby reducing overall computational burden while maintaining allocation efficiency.
3Quantity of substance
If conventional storage-only allocation is used, then capacity requirements are met, but network path bandwidth and switch reliability are insufficient
Solution Approach 1:
The patent introduces network resource optimization as an intermediary layer between storage allocation and application performance. The system optimizes network paths and switch configurations to ensure sufficient bandwidth and reliability for data transmission between storage resources and applications, thereby resolving the contradiction between meeting storage capacity requirements and ensuring network path reliability.
Data Source
AI summary
A computer-implemented method for determining, from a system including a plurality of data center resources, at least one configuration of data center resources for an implementation of an application. The method includes receiving application information and receiving information regarding known internal features up the data center resources. The method also includes provisioning the system of data center resources and creating possible configurations of data center resources for implementing application. The method also includes correlating models and data center resources to create an interrelated representation of the models and the data center resources. The models predict a relationship of parameters for the possible configurations. The method also includes creating a multiple dimensional analysis of parameters for the possible configurations of data center resources using the interrelated representation and selecting a configuration of data center resources from the possible configurations using the multiple dimensional analysis of parameters.