Resource Estimation for Virtualized Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Evaluating resource sufficiency for complex business-level operations in critical enterprise applications is challenging due to their impact on multi-server environments and service level agreements, especially when determining resource allocation for concurrent users.
Innovation Solution
An analysis application identifies servers supporting application operations, records resource utilization, and creates linear equations to estimate resource costs of each operation type, enabling fine-tuning and predicting resource needs for concurrent users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If benchmarks on simple atomic system operations are used for evaluation, then the evaluation process is simple and easy to implement, but the evaluation results do not reflect the complexity of business level operations and their impact on the application environment
Solution Approach 1:
The patent segments business level operations into hierarchical components: business level operations are broken down into screen level operations, which are further segmented into atomic system operations. This segmentation allows the system to evaluate complex business operations by analyzing their constituent parts while maintaining the ability to assess overall business impact.
Solution Approach 2:
The patent introduces screen level operations as intermediary elements between atomic system operations and business level operations. These screen operations serve as mediators that capture the complexity of user interactions while providing a structured layer for analysis, bridging the gap between simple atomic operations and complex business processes.
2Productivity
If resource allocation is determined for each application operation type, then resource utilization can be optimized, but the system complexity increases due to virtual provisioning and parallel load operations
Solution Approach 1:
The patent creates a universal evaluation framework that works across multiple server types and virtual provisioning scenarios. The system uses a standardized approach to model and evaluate resource requirements that can be applied uniformly across different server configurations, virtualization environments, and operation types, reducing the complexity of managing diverse systems.
Solution Approach 2:
The patent transforms the complex problem of resource allocation by changing parameters from direct monitoring of individual virtual resources to mathematical modeling of operation-resource relationships. By using linear equations and statistical analysis, the system converts complex virtual provisioning parameters into manageable resource cost estimates that can be optimized without directly managing system complexity.
3Measurement precision
If the impact of each business level operation on the entire application environment is analyzed, then accurate resource requirements can be determined, but the analysis becomes computationally intensive and time-consuming
Solution Approach 1:
The patent performs preliminary analysis by pre-establishing the hierarchical relationship between business level operations, screen operations, and atomic operations. The system pre-defines the mapping between operations and their resource requirements, creating a ready-to-use evaluation framework that can quickly assess resource needs without performing exhaustive analysis each time an operation is evaluated.
Solution Approach 2:
The patent replaces complex mechanical analysis of operation impacts with mathematical modeling. Instead of physically tracing each operation's impact through the system, the patent uses linear equations and statistical methods to calculate resource requirements, substituting computational mathematics for exhaustive system analysis and significantly reducing evaluation time.
Data Source
AI summary
Estimating required resources to support a specific number of users in a virtually provisioned environment is described. Servers are identified that support application operations associated with executing an application, based on a configuration file. A count of each type of application operation currently executing and a resource utilization associated with each of the servers are recorded. A set of linear equations is created and solved to estimate resource costs of each individual type of application operation and ultimately to calculate required resource costs to support the projected number of concurrent users.