Effective Centroid Calculation for Data Center Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online conferencing systems face inefficiencies in selecting the optimal data center to host conference calls, as they typically rely solely on geographical distance without considering participant device types, connection types, and relative importance, leading to suboptimal performance and quality.
Innovation Solution
A method for selecting an initial data center based on participant locations and effectively calculating an 'effective centroid' that incorporates device type, connection type, and weighting factors to determine the best data center for hosting conference calls, with the option to hand off to a different data center if conditions change during the call.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data center selection is based solely on geographical distance, then the selection process is simple, but the conference call quality and performance are suboptimal
Solution Approach 1:
The patent transforms the single parameter of geographical distance into multiple parameters including physical distance, effective distance factor, device type, connection type, and participant importance weighting. This multi-parameter approach enables more accurate data center selection that optimizes conference call quality while managing complexity through systematic evaluation
Solution Approach 2:
The patent introduces an intermediary calculation layer called 'effective distance' that mediates between raw geographical distance and final data center selection. This intermediary incorporates device characteristics, connection types, and participant importance to bridge the gap between simple distance metrics and complex quality requirements
2Productivity
If data center selection considers multiple factors like device type and connection type, then conference call performance is improved, but the calculation complexity increases
Solution Approach 1:
The patent changes the selection criteria from simple geographical distance to a composite metric incorporating device type, connection type, and participant importance. This transformation improves selection efficiency by choosing data centers that better match participant characteristics, though it increases algorithmic complexity
Solution Approach 2:
The patent performs preliminary calculations of effective distance factors for each participant before making the final data center selection. This preliminary action organizes complex data into structured metrics, making the subsequent selection process more efficient despite the increased complexity of individual calculations
3Reliability
If the system continuously monitors and recalculates optimal data center, then conference call quality is optimized, but computational overhead increases
Solution Approach 1:
The patent implements a dynamic data center selection system that can recalculate optimal data centers based on changing participant conditions. The system balances quality optimization with resource consumption by performing recalculations only when necessary, adapting to real-time changes in participant locations, device types, and connection types
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
An initial data center is selected to host an online conference. This data center can be selected based on the geographical locations of the participants. Typically, the data center closest to the centroid of the participants is selected. During (or before) the conference call, an 'effective' centroid is calculated based on effective distances. Effective distances are based on a combination of the physical distance between a participant and a data center, and an effective distance factor (or weighting) that is based on one or more of the participant's device/driver type, the participant's network type, the participant's connection type, and a participant weighting factor.