Edge-Based Bidding for Efficient Edge-to-Cloud Data Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current edge-cloud infrastructure lacks an optimized scheme for a remote compute device to request data sets from the appropriate edge compute devices, which hinders efficient AI-based data analysis.
Innovation Solution
Implementing an edge-based bidding process where a remote compute device sends data bid requests to multiple edge compute devices, evaluates their bidding values based on factors like workload load, subsystem health, and data transmission cost, and selects the most suitable device for data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a remote compute device requests data from multiple edge compute devices without an optimized selection scheme, then data availability is improved, but data transmission efficiency deteriorates
Solution Approach 1:
The system performs preliminary actions by having edge compute devices calculate and submit bidding values before the actual data transfer occurs. The remote compute device receives and evaluates these pre-calculated bidding values to determine the optimal target edge device, avoiding the need to evaluate multiple devices during the actual data transfer process.
Solution Approach 2:
Edge compute devices autonomously calculate their own bidding values based on their current workload, subsystem health, and data transmission cost estimates. Each device independently evaluates its capability to serve the data request and communicates this self-assessment to the remote compute device, enabling decentralized decision-making.
2Measurement precision
If edge compute devices continuously report detailed status information, then system optimization accuracy is improved, but communication overhead increases
Solution Approach 1:
The system allows each edge compute device to report its specific local conditions (workload load, subsystem health status, data transmission cost) rather than requiring uniform comprehensive status reporting from all devices. This enables the remote compute device to make optimized decisions based on relevant local qualities of each edge device.
Data Source
AI summary
A remote compute device stores a data collection policy. A processor receives a data request from an application and determines a plurality of data sets within the data request. The processor provides a data bid request to multiple edge compute devices. The processor receives multiple bidding values from the edge compute devices. A different one of the bidding values corresponds to a different one of the edge compute devices. Based on the multiple bidding values and the data collection policy, the processor determines a target edge compute device. The processor provides a data request to the target edge compute device. The processor receives data from the target edge compute device and provide the data to the application.