Virtual Environment Resource Prediction Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting resource usage in virtual environments fail to accurately account for virtualization overheads, leading to potential deployment of applications on servers with insufficient resources, resulting in suboptimal performance.
Innovation Solution
A methodical approach using a benchmark suite and regression-based computation to create a prediction model that maps native hardware system resource usage profiles to virtual environment profiles, accounting for CPU, network, and disk-intensive workloads to estimate resource requirements in a virtualized environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing prediction methods are used to estimate resource usage in virtual environments, then the prediction process is simple and quick, but the accuracy of resource usage prediction is insufficient due to failure to account for virtualization overheads
Solution Approach 1:
The patent applies preliminary action by pre-calculating virtualization overhead factors through benchmarking before actual application deployment. The system performs advance measurements of CPU, memory, storage, and network overheads using representative workloads, then uses these pre-computed factors to adjust resource usage predictions. This approach improves prediction accuracy without requiring complex real-time calculations during application deployment.
Solution Approach 2:
The patent introduces an intermediary prediction model that acts as a mediator between simple resource monitoring and complex performance analysis. This model incorporates virtualization overhead factors as intermediate adjustment parameters, transforming basic resource usage metrics into more accurate predictions by applying overhead multipliers derived from benchmarking data.
2Productivity
If server consolidation is implemented to reduce server sprawl, then the number of servers is reduced and space is saved, but resource allocation accuracy deteriorates due to insufficient accounting for virtualization overheads
Solution Approach 1:
The patent implements feedback by continuously monitoring actual resource consumption in virtualized environments and comparing it against predicted values. The system uses this feedback to refine overhead factor estimates and improve future predictions. Benchmarking results are periodically updated based on actual performance data, creating a closed-loop system that enhances both consolidation efficiency and allocation reliability.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting resource allocation parameters based on virtualization overhead factors. Instead of using fixed allocation ratios, the system modifies CPU, memory, storage, and network allocation parameters according to measured overheads from benchmarking, enabling more accurate resource provisioning in consolidated server environments.
3Loss of time
If virtualization overheads are not accounted for in resource predictions, then the deployment process is faster and simpler, but applications may be deployed on servers with insufficient resources leading to suboptimal performance
Solution Approach 1:
The patent applies preliminary action by pre-computing virtualization overhead factors through benchmarking before application deployment. This advance preparation enables fast deployment decisions without sacrificing accuracy, as the overhead adjustment parameters are already available from previous benchmarking runs rather than requiring real-time calculation during deployment.
Solution Approach 2:
The patent applies partial action by focusing overhead measurements on the most critical resource parameters (CPU, memory, storage, network) rather than attempting to measure all possible system metrics. This selective approach provides sufficient accuracy for deployment decisions while maintaining fast processing speeds.
Data Source
AI summary
Described herein is a system for predicting resource usage of an application running in a virtual environment. The system comprises a first hardware platform implementing a native hardware system in which an application natively resides and executes, the native hardware system operating to execute a predetermined set of benchmarks that includes at least one of: a computation-intensive workload, a network-intensive workload, and a disk-intensive workload; a second hardware platform implementing a virtual environment therein, the virtual environment operating to execute the predetermined set of benchmarks; a third hardware platform operating to collect first resource usage traces from the first hardware platform and second resource usage traces from the second hardware platform; wherein the third hardware platform further operating to generate at least one prediction model that predicts a resource usage of the application executing in the virtual environment based on the collected first and second resource usage traces.