high performance computing on graphics processing units: hgpu.org

Smartphone users often replace their devices prematurely for newer models, contributing to the growing issue of waste electrical and electronic equipment (WEEE). Repurposing these devices to extend their life cycle by assigning them new roles can help mitigate this problem. This thesis explores the feasibility of creating a cluster using upcycled smartphones deployed with the K3S framework. We developed an image classification application capable of handling HTTP requests, utilizing the TensorFlow Lite and Crow frameworks. The devices, operating system, and framework were adapted for compatibility. Preliminary networking evaluations indicate that these devices can effectively be part of a cluster across various networking mediums. Performance evaluations of the image classification application running on this cluster indicate that performance scales effectively up to a medium-specific threshold with Ethernet and Wi-Fi 5.0GHz, confirming that upcycled devices are viable for a K3S cluster design aimed at low-cost edge computing.