A Distributed Deep Meta Task Using Internet of Things

The principal object of the present disclosure is to provide a novel strategy for offloading work from latency-sensitive IoT applications while maximizing resource utilization in Edge Clouds. This unique innovation is provided to introduce a selection of concepts in a simplified form to be further described.

Accordingly, in order to maximize resource utilization in Edge Clouds, the work aims to provide a novel approach for offloading work from latency-sensitive IoT applications. The suggested method is also used to plan task offloading, which cuts down on overall service times and makes the most of available resources.

Additionally, the present invention describes an Edge Clouds architecture based on DLTs that includes the elements required to support the scheduling of IoT application task offloads. In addition to resource utilizations, heterogeneities, and total times, the present invention reduces characteristics of latency-sensitive applications such as CPU energies, executions, network demands, and delays. A novel method called Distributed Deep Meta learning-driven Task Offloading (DDMTO) is presented that combines a meta-algorithm and deep neural networks (DNNs).