Dynamic Cloud Deployment Metrics Engine
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The lack of transparency and uniformity in mapping virtual resources to physical resources in cloud computing infrastructures leads to performance variability and increased costs for users, making it challenging to select optimal cloud configurations that balance performance and cost.
Innovation Solution
A system and method for dynamically and adaptively deploying applications on distributed computing systems by obtaining real-time metrics and using a Price-Performance Monitor and Recommendation Engine to select cloud configurations based on user-specified criteria, optimizing performance while minimizing deployment costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If virtual resources are mapped to physical resources in cloud infrastructures, then application deployment is enabled, but performance variability and cost uncertainty increase due to lack of transparency and uniformity
Solution Approach 1:
The system continuously collects real-time metrics from cloud configurations and uses this feedback to dynamically update performance models. The recommendation engine leverages this feedback loop to provide increasingly accurate predictions about cloud configuration performance, enabling users to make reliable selections despite the inherent variability in virtual-to-physical resource mapping.
Solution Approach 2:
The patent replaces manual cloud configuration selection with an automated recommendation engine that uses machine learning models. This substitution transforms the mechanical process of browsing and selecting cloud configurations into an intelligent system that predicts performance based on historical data and real-time metrics, thereby reducing performance variability in the selection process.
2Measurement precision
If real-time metrics are collected from multiple cloud configurations, then optimal cloud selection is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The system extracts only the most relevant real-time metrics from the vast amount of available cloud configuration data. By identifying and focusing on key performance indicators that directly impact application performance and cost, the system reduces the complexity of data collection and processing while maintaining high measurement precision for cloud configuration selection.
Solution Approach 2:
The system performs preliminary processing of metrics data by pre-filtering, normalizing, and storing it in a structured format. This preliminary action reduces the complexity of real-time processing by having data ready in an optimized structure, allowing the recommendation engine to quickly query and analyze only the necessary metrics without processing the entire raw data set.
3Productivity
If dynamic cloud configuration selection is implemented, then cost optimization is achieved, but deployment time and computational overhead increase
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing performance models for various cloud configurations based on historical data. When a deployment is needed, the recommendation engine can quickly query these pre-computed models rather than performing complex analyses in real-time, significantly reducing configuration selection time while maintaining cost optimization.
Solution Approach 2:
The system uses partial action by evaluating only the most relevant cloud configurations based on user-specified criteria and application requirements. Rather than analyzing all possible cloud configurations exhaustively, the system identifies and evaluates a subset of likely candidates using the performance models, reducing computational overhead and deployment time while achieving sufficient cost optimization.
Data Source
AI summary
Embodiments of apparatus, systems and methods facilitate the adaptive deployment of a distributed computing application on at least one selected cloud configuration from a plurality of cloud configurations based on dynamically obtained and/or compiled metrics pertaining to the cloud configurations and to the distributed computing application and/or user specified criteria pertaining to the metrics. In some embodiments, an infrastructure independent representation of the distributed computing application is adapted to the selected cloud configuration and run on the selected cloud configuration by utilizing a cloud-specific implementation of the infrastructure independent representation of the distributed computing application. The cloud-specific implementation of the infrastructure independent representation corresponds to the cloud infrastructure on which the distributed application is run.


