Container Deployment via Standardized Performance Inquiry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The deployment of containers on cloud providers is inefficient due to the difficulty in determining the optimal provider based on performance metrics, which vary across providers and are often represented differently, making direct comparisons infeasible.
Innovation Solution
A central management system negotiates container deployment using an information exchange protocol with preconfigured inquiry and response formats to request and provide performance data, allowing containers to be deployed on best-fit cloud providers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If performance metrics are collected from multiple cloud providers, then deployment quality improves, but data heterogeneity makes comparison infeasible
Solution Approach 1:
The patent applies homogeneity by standardizing heterogeneous performance metrics from different cloud providers into a unified data structure. The system normalizes varying data formats, units, and schemas into a common representation, enabling direct comparison of performance metrics across providers while maintaining the ability to handle diverse input formats.
Solution Approach 2:
The patent introduces an intermediary normalization layer that mediates between diverse cloud provider data formats and the deployment decision-making process. This intermediary component transforms and harmonizes incoming performance data before comparison, acting as a bridge that enables feasible metrics comparison without requiring changes to the underlying cloud providers' data structures.
2Productivity
If real-time performance data is processed, then deployment efficiency improves, but data processing complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-defining normalization rules, data transformation templates, and comparison criteria before performance data arrives. The system prepares the processing framework in advance, so when real-time metrics stream in, they can be quickly normalized and compared without ad-hoc processing complexity, thereby maintaining deployment efficiency.
Solution Approach 2:
The patent transforms the complexity issue into a parameter management problem by changing the state of performance data from heterogeneous to homogeneous through systematic parameter transformation. The system applies consistent transformation parameters and normalization factors to convert varying data formats into a standard representation, reducing processing complexity while enabling real-time analysis.
Data Source
AI summary
In one embodiment, a method includes receiving a trigger to deploy a particular container on cloud resources accessible thereto such that the cloud resources are provided by a plurality of cloud providers and such that the computer system and the plurality of cloud providers are configured to negotiate container deployment using an information exchange protocol. The information exchange protocol includes a preconfigured inquiry format and a preconfigured inquiry-response format. The method further includes generating a performance inquiry in relation to the particular container. In addition, the method includes transmitting the performance inquiry to the plurality of cloud providers. Moreover, the method includes receiving inquiry responses from at least some of the plurality of cloud providers. Additionally, the method includes causing the particular container to be deployed on resources of the particular cloud provider.