Drift-Based Dynamic Clusters for IoT Load Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The exponential growth of IoT devices leads to a significant increase in data exchange chatter, resulting in astronomical amounts of packet communications, which can be reduced by determining the geographical location of devices and using spare compute cycles for workload processing, thereby minimizing data communications.
Innovation Solution
A method for distributed load processing using drift-based dynamic clusters of IoT devices, where a central IoT device selects and configures nearby devices to form clusters, processes workloads, and dynamically replaces devices based on performance metrics and resource availability to optimize workload distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of interconnected IoT devices is increased to offer more services and devices, then the service coverage and device availability are improved, but the data exchange chatter and packet communications increase astronomically
Solution Approach 1:
The patent segments the IoT network into geographic clusters, where each cluster independently processes local workloads. This segmentation reduces the need for global packet communications by confining data exchange to local cluster boundaries, thereby maintaining service coverage while reducing overall communication volume.
Solution Approach 2:
The patent implements local quality by enabling workloads to be processed where the data resides (at the edge devices and local clusters) rather than centralizing all processing. This local processing approach minimizes data exchange chatter by eliminating unnecessary remote communications while preserving service availability.
2Device complexity
If traditional centralized processing is used for workload distribution, then processing control is simplified, but data communications and packet transmissions increase significantly
Solution Approach 1:
The patent divides the centralized processing architecture into distributed geographic clusters, each with local processing capabilities. This segmentation maintains simplified control within each cluster while reducing overall data communications by eliminating the need for all data to traverse to a central processing point.
Solution Approach 2:
The patent introduces a geographic dimension to workload processing by organizing devices into location-based clusters. This dimensional change allows workloads to be processed locally near the data source, dramatically reducing data communication distances and volumes while maintaining processing control through the cluster management system.
3Quantity of substance
If IoT devices are selected based on geographical location proximity to data sources, then data communications are minimized, but device selection and cluster formation complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-organizing IoT devices into geographic clusters based on their locations relative to data sources. This pre-organization simplifies subsequent workload distribution decisions, as the clustering structure is established in advance, reducing the complexity of real-time device selection while maintaining minimal data communication benefits.
4Reliability
If dynamic cluster formation and device replacement is implemented based on performance metrics, then workload processing reliability is improved, but system complexity and monitoring overhead increase
Solution Approach 1:
The patent implements feedback mechanisms by continuously monitoring device performance metrics and using this information to dynamically adjust cluster composition and workload distribution. This feedback loop improves reliability by ensuring workloads are processed by capable devices while managing system complexity through automated monitoring and adjustment protocols.
Solution Approach 2:
The patent introduces dynamics by enabling clusters to adapt their composition over time based on device performance, availability, and workload requirements. This dynamic approach improves reliability by automatically replacing underperforming devices while managing complexity through systematic replacement protocols rather than static configurations.
Data Source
AI summary
For distributed processing using drift-based dynamic clustering of Internet of Things (IoT) devices, at a central device, a data source to be used for processing a workload is determined. A set is selected of devices operating within a threshold distance from the data source at a first time. A first subset of the set of devices is selected to form a cluster of devices. Each device in the first subset satisfies a clustering condition. A first device in the first subset is instructed to configure an application at the first device to participate in the cluster and process the workload. From a performance check on the first device, a change is discovered in a performance metric. In response to the change resulting from an increased demand for a computing resource at the first device, the first device is replaced with a second device from the first subset.