IoT Application Migration for Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lightweight IoT devices, such as drones, lack the necessary computational power and experience high latency in data processing, making it difficult to perform resource-intensive tasks like image processing and live-navigation due to their inability to carry large power sources and storage/memory/CPU elements.
Innovation Solution
Implementing a system that automatically migrates compute and memory resources to the IoT device's location using mini data centers, pre-positioning resources, code, and data through replication solutions like CMotion and Dell EMC RecoverPoint, ensuring low latency and enabling resource-intensive operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If lightweight IoT devices are used, then device portability and ease of operation are improved, but computational power and data processing capability deteriorate
Solution Approach 1:
The patent introduces a cloud-based computing intermediary that mediates between the lightweight IoT device and the computational resources. The device offloads data processing tasks to cloud servers, which act as intermediaries providing the necessary computational power without requiring the device itself to have high processing capability. This resolves the contradiction by separating the device's portability function from the computational function.
Solution Approach 2:
The patent moves computational resources from the physical dimension (on-device hardware) to the network dimension (cloud infrastructure). By transitioning computation from local CPU to remote servers connected via network, the system achieves high computational power while maintaining device lightness. This dimensional shift resolves the contradiction between device weight and processing capability.
2Power
If computations are performed in the cloud, then computational power is improved, but data processing latency increases
Solution Approach 1:
The patent implements predictive pre-positioning of data and computational resources. The system predicts future device locations and pre-loads necessary data to edge servers in those locations before the device arrives. This preliminary action reduces latency by ensuring data is already available at the edge when the device needs it, rather than fetching from remote cloud centers.
Solution Approach 2:
The patent creates locally optimized computing environments at edge servers near the device. Instead of uniform remote cloud processing, the system establishes local computation nodes that provide fast, low-latency processing. This local quality approach ensures computational power is delivered with minimal latency by positioning resources geographically close to the device.
3Productivity
If more computational resources are provided locally, then processing capability is improved, but device weight and power consumption increase
Solution Approach 1:
The patent extracts computational resources from the mobile device itself and relocates them to external cloud and edge infrastructure. By taking out the heavy CPU, storage, and power requirements from the device, the system maintains lightweight portability while achieving high processing capability through network-connected resources. This extraction resolves the contradiction by separating device weight from processing power.
Solution Approach 2:
The patent creates virtual copies of computational environments at edge servers that mirror the device's local execution context. Instead of moving physical hardware resources to the device, the system creates virtualized copies of the computational environment remotely, allowing the lightweight device to access full processing capability through virtualization without carrying actual hardware resources.
Data Source
AI summary
One example method includes receiving input concerning a mobile IoT device, and the input includes information about a location of the mobile IoT device, information about whether the mobile IoT device is moving, and, when the mobile IoT device is moving, information about the range, speed, and bearing of the mobile IoT device. Next, the method includes generating a predicted location of the mobile IoT device based on the inputs received, using the predicted location of the mobile IoT device and a map of nodes in an environment where the mobile IoT device is located to make a migration decision concerning an application used by the mobile IoT device, and migrating the application from a present location to a node expected to be accessible by the mobile IoT device when the mobile IoT device reaches the predicted location, and the node and present location are physically separated by a distance.


