PERISAI: Design and Implementation of a Remote Deployment System for IoT Applications
Muhammad Garebaldhie ER Rahman, Muhammad Zuhri Catur Candra · 2024
As the use of Internet of Things (IoT) devices continues to expand, the management of these devices has become increasingly complex and time-consuming. One of the key processes in managing IoT devices is deployment, which involves installing and updating software on the devices. Traditionally, this process has been performed manually, device by device, often through a wired connection such as USB, which is highly inefficient and labor-intensive. With advancements in technology, Over the Air (OTA) methods have emerged, allowing for wireless deployment of software updates. However, these methods still face significant challenges, particularly in terms of scalability and managing a diverse range of devices with different capabilities and constraints. Issues such as device heterogeneity, difficulty in diagnosing devices remotely, hardware resource limitations, and security concerns remain prevalent. This research addresses these challenges by designing and implementing PERI SAI, a remote deployment system specifically tailored for IoT applications. PERISAI is designed to be platform-agnostic, meaning it can operate on any platform as long as it can run Kubernetes, and is optimized for devices with limited resources. The system utilizes Kubernetes, a widely adopted container orchestration tool, with the lightweight K3s distribution to manage deployments efficiently on resource-constrained devices such as Raspberry Pi. PERISAI has been tested by deploying a blink LED program across two Raspberry Pi devices with a single click, achieving successful deployment in under 4.7 seconds.