• cloud computing
• edge computing
• energy efficient scheduler
• faas
• function scheduling
• milp

Future edge-to-cloud continuum platforms adopting the Function-as-a-Service (FaaS) paradigm will require energy management mechanisms in order to minimize their energy consumption when deployed on a large-scale. Addressing this issue poses several challenges including predicting workload intesity to dynamically adjust computing resources and how to organize the execution of functions efficiently within computing nodes. This thesis aims at designing an energy-efficient autoscaling mechanism that keeps the energy consumption in an edge zone low by having the fewest number of active nodes capable of handling the current workload. This will be carried out specifically considering the trade-off between energy utilization and the Quality of Service requirements of the functions. The solution will be designed and implemented specifically in Serverledge, a FaaS platform designed for the Edge-to-Cloud continuum.”