Workload Profile Generation for Converged Infrastructure Resource Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale IT organizations face challenges in understanding and managing the characteristics and behaviors of virtualized computing components in relation to physical resources, requiring time-consuming and labor-intensive coding to optimize resource utilization.
Innovation Solution
A system and method for generating workload profiles of physical and logical computing components to perform predictive workload analysis, identifying optimal resources for executing client resources by monitoring and processing utilization data across a converged infrastructure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If virtualization and cloud computing techniques are implemented to provide computing services efficiently, then service delivery efficiency is improved, but understanding and managing the characteristics and behaviors of virtualized components in relation to physical resources becomes complex and labor-intensive
Solution Approach 1:
The patent introduces a workload profile generation system that acts as an intermediary between virtualized computing components and physical resources. This system automatically generates workload profiles that characterize and predict the behavior of virtual components, simplifying the management complexity while maintaining the efficiency benefits of virtualization and cloud computing.
2Adaptability or versatility
If numerous applications are run on hundreds or thousands of virtual components across globally disparate physical resources, then service coverage and scalability are improved, but time-consuming and labor-intensive coding is required to understand component characteristics and behaviors
Solution Approach 1:
The patent implements preliminary action by automatically generating workload profiles that pre-characterize the behavior and characteristics of virtualized computing components before they are deployed or scaled. This preliminary characterization eliminates the need for time-consuming manual analysis when scaling to hundreds or thousands of virtual components across global infrastructure.
3Measurement precision
If manual analysis and coding are used to understand virtualized component behaviors, then detailed understanding of component characteristics is achieved, but resource allocation optimization becomes labor-intensive and slow
Solution Approach 1:
The patent replaces the manual mechanical process of analyzing and coding to understand virtualized component behaviors with an automated system that generates workload profiles using predictive analytics. This substitution maintains precise understanding of component characteristics while dramatically improving the speed of resource allocation optimization.
Data Source
AI summary
Methods and/or systems for performing workload analysis within an arrangement of interconnected computing devices, such as a converged infrastructure, are disclosed. A prediction system may generate a workload associated with physical and/or logical components of the converged infrastructure that are utilized to execute a client resource. The prediction system may monitor the utilization behavior of the various logical and/or physical components associated with the workload over a particular period of time to generate a workload profile. Subsequently, the prediction system may execute a prediction workload analysis algorithm that accesses the workload profile to identify optimal physical resources in the converged infrastructure that may be available to execute other workloads.