With the high increase of internet-connected devices, fog computing has attracted great attention in recent years. Fog computing is geographically distributed computing extends cloud computing provides ubiquitous connected heterogeneous devices, storage, networking, and application services in a highly visualized platform at the edge of the network. Fog and cloud enable several IoT devices and applications with low latency because fog significantly manages the massive amount of data generated from IoT devices and improves the service latency based on resources closer to the edge.
Conventionally, scheduling employs allocating tasks to capable resources at a specific time. The existing works are based on distributing event-based across edge and cloud resources to support IoT applications for latency. Though, the placement schedule needs to meet constraints such as the throughput capacity supported on edge and cloud machines by the queries, bandwidth and latency limits of the network, and energy capacity of the edge devices. Thus, the deep learning approach considers latency sensitivity for IoT applications and effectively detects scheduling problems involving resources in both fog and cloud.