Cloud Resource Evaluation Using Live-Condition Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud-based platforms face challenges in accurately evaluating resource requirement recommendations due to reliance on historical data, leading to inefficient resource utilization, excessive energy usage, and increased operating costs, as predictive models fail to capture the dynamic and unpredictable nature of live deployments.
Innovation Solution
A simulated computing environment is generated using current system characteristics and activity data from live deployments to realistically evaluate resource requirement recommendations, incorporating a simulated scheduler that mimics the behavior of the orchestration system, allowing for thorough analysis of resource utilization and performance across various scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If historical data is used to evaluate resource requirement recommendations, then the evaluation process is simple, but the accuracy of resource utilization prediction deteriorates due to inability to capture dynamic live deployment conditions
Solution Approach 1:
The patent creates a simulated computing environment that copies the structure and behavior of the live cloud platform, including simulated nodes, containers, and orchestration systems. This allows accurate evaluation of resource recommendations without directly impacting the live system, resolving the contradiction by providing a faithful replica that captures dynamic conditions while maintaining evaluation isolation.
Solution Approach 2:
The system performs preliminary evaluation of resource requirement recommendations in the simulated environment before they are applied to the live cloud platform. By pre-testing recommendations using sampled live deployment data and measuring metrics like resource utilization and performance, the system avoids inaccurate predictions that would result from using only historical data, while establishing a structured evaluation framework.
2Productivity
If resource requirement recommendations are applied without thorough evaluation, then deployment speed is fast, but resource utilization efficiency deteriorates leading to excessive energy usage and increased operating costs
Solution Approach 1:
The system performs preliminary evaluation of resource recommendations in the simulated environment before live deployment, measuring resource utilization metrics and performance indicators. This pre-evaluation step ensures that only well-optimized recommendations are applied to the live system, improving energy efficiency without significantly delaying deployment through automated simulation execution.
Solution Approach 2:
The simulated computing environment autonomously executes the pending software deployment and automatically measures resource utilization and performance metrics without requiring manual intervention. The system self-evaluates recommendations by comparing actual resource consumption against recommendations, identifying inefficiencies in energy usage and resource allocation independently.
3Measurement precision
If a simulated computing environment is generated and used for evaluation, then resource utilization accuracy is improved, but system complexity increases
Solution Approach 1:
The patent creates a simulated computing environment that copies the essential structure and behavior of the live cloud platform, including simulated nodes, containers, and orchestration systems. This allows accurate evaluation of resource recommendations while maintaining a manageable simulation scope that focuses on critical components rather than replicating the entire live system.
Solution Approach 2:
The simulated computing environment acts as an intermediary between the resource recommendation system and the live cloud platform. It provides accurate measurement of resource utilization metrics without directly complexity the live system, serving as a buffer that captures dynamic conditions while isolating the production environment from evaluation operations.
4Reliability
If multiple test scenarios are generated and evaluated, then evaluation robustness is improved, but evaluation time increases
Solution Approach 1:
The system generates multiple test scenarios with modified activity datasets to evaluate resource recommendations under varying conditions, improving robustness. However, it applies partial action by selecting a representative subset of scenarios rather than exhaustively testing all possible variations, balancing evaluation thoroughness with acceptable time constraints through targeted scenario selection.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
The techniques disclosed herein enable a system to perform a robust evaluation of resource requirement recommendations through a simulated computing environment that closely resembles current conditions of a live computing environment. To achieve this, system characteristics such as CPU, RAM, and storage are extracted from currently available computing resources at the live computing environment. In addition, active software deployments at the live computing environment are randomly sampled to generate an activity dataset. The system characteristics and the activity dataset are then used to generate the simulated computing environment. Instances of a pending software deployment are then assigned to the simulated computing environment according to a resource requirement recommendation. The instances are then executed across various scenarios and analyzed to calculate a level of resource utilization. Consequently, several resource requirement recommendations can be evaluated and compared simultaneously thereby enabling the system to select the best resource requirement recommendation.