ABSTRACT: As the Internet of Things (IoT) and cloud technologies advance, a growing number of IoT devices and sensors send massive amounts of data to cloud data centres for processing. While providing us with a great deal of convenience, cloud-based computing and storage also introduces a slew of security issues, such as data gathering misuse and cloud-based web servers. Traditional intrusion detection systems (IDS) and web application firewalls (WAF) are becoming incompatible with the new network environment, prompting the development of similar solutions based on machine learning or deep learning. However, because data centralization is a more appealing reward, cloud-IoT systems boost attacks against web servers. In this research, we propose a web assault detection system based on distributed deep learning that takes advantage of URL analysis. The system, which is implemented on edge devices, is meant to identify web attacks. In the Edg paradigm, the cloud tackles the aforementioned difficulties. To improve the system's stability and update convenience, many concurrent deep models are deployed. Using two concurrent deep models, we ran trials on the system and compared it to other systems using a variety of datasets. The testing findings show that the system is competitive in identifying web attacks, with accuracy of 99.410 percent, TPR of 98.91 percent, and DRN of 99.55 percent.Keywords— Distributed Deep Learning, Distributed System, Edge of Things, Web Attack Detection