Cloud Resource Placement via Adaptive Rule Sets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed computing systems, users with varying levels of expertise face challenges in making optimal decisions regarding the deployment of computing resources, as not all users have the same abilities or understanding to customize their services effectively, leading to suboptimal placement of data centers and resource allocation.
Innovation Solution
A service provider network that assists users in selecting physical locations for computing resources by collecting parameters such as cost tolerance, latency, and fault risk diversity, using rule sets and a learning algorithm to suggest optimal locations, and enabling users to input parameters through a user interface to create custom selection criteria for deploying computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If users are empowered to make decisions about computing resource placement, then service customization capability is improved, but decision quality deteriorates due to varying user expertise levels
Solution Approach 1:
The patent introduces an intermediary system that acts as a mediator between users and the complex decision-making process for resource placement. This intermediary provides automated recommendations based on multiple parameters (cost, latency, fault risk diversity) and allows users to either accept these recommendations or make informed adjustments, thus bridging the gap between user empowerment and decision quality.
Solution Approach 2:
The system implements feedback mechanisms that provide users with information about the implications of their decisions on cost, latency, and fault risk diversity. This feedback loop enables users to understand the consequences of their choices and make more informed decisions, improving decision quality while maintaining customization capability.
2Measurement precision
If automated systems make resource placement decisions, then decision quality is improved, but user control deteriorates
Solution Approach 1:
The system dynamically adjusts the level of automation based on user preferences and expertise. Users can choose to fully accept automated recommendations, partially review and modify them, or have more hands-on control. This dynamic approach allows the system to optimize decision quality while adapting to varying user control requirements.
3Productivity
If multiple parameters are considered for resource placement, then placement optimization is improved, but system complexity deteriorates
Solution Approach 1:
The patent segments the complex multi-parameter optimization problem into distinct, manageable components: cost analysis, latency assessment, and fault risk diversity evaluation. Each parameter is evaluated separately using specific algorithms, and the results are integrated to form comprehensive recommendations. This segmentation reduces system complexity by breaking down the overall problem into smaller, more tractable sub-problems.
Data Source
AI summary
A service provider may select one or more physical locations of existing data centers for the user for deployment of the user's computing resources. In various embodiments, the service provider may collect parameters from the user to create a custom selection of physical locations for computing services to be deployed for the user, which may be a strategically selected subset of all of the existing data centers available. Some parameters for selection may include expected location of end-users, cost tolerance, latency tolerance, and fault risk diversity. In some situations, cost may have an inverse relationship with latency, such that as cost increases, latency decreases (and vice versa). The rule sets may be created using the parameters, possibly with weights assigned to different parameters based on information received via the user interface. The rule sets may be formed and used to select physical locations for deployment of computing resources.