NECSTFridayTalk – Runtime Deep Learning Orchestration for Heterogeneous Edge Platforms
Events

NECSTFridayTalk – Runtime Deep Learning Orchestration for Heterogeneous Edge Platforms

NOVEMBER 20, 2026

Featured image 1

November 20th, 2026 | 11:30 am
DEIB - NECSTLab Meeting Room (Bld. 20)
Online by Zoom

Contact: Prof. Marco Santambrogio

Abstract

On Friday, November 20th, 2026, we will have a new talk for the series #NECSTFridayTalk.

During this talk, we will have, as speaker, Antonio Miele, Professor at Dipartimento di Elettronica, Informazione e Bioingegneria.

Over the past decade, the demand for lower latency, energy efficiency, and privacy has shifted deep learning inference from the cloud to the edge. Modern edge systems comprise heterogeneous devices, including CPUs, GPUs, and specialized accelerators, handling dynamic workloads. This raises key orchestration challenges: mapping tasks to devices, selecting models, sharing resources, configuring hardware, and meeting latency, accuracy, and energy constraints. The variability of workloads and platforms further calls for adaptive runtime management. In this talk, we review recent advances in deep learning orchestration for heterogeneous edge systems, drawing on my research conducted in collaboration with a team at the University of Turku. We focus on adaptive scheduling, model selection, and runtime resource management techniques that enable efficient inference under dynamic workload and resource constraints.

The NECSTLab is a DEIB laboratory, with different research lines on advanced topics in computing systems: from architectural characteristics, to hardware-software codesign methodologies, to security and dependability issues of complex system architectures.
#NECSTLab #Computerscience 

Every week, the “NECSTFridayTalk” invites researchers, professionals or entrepreneurs to share their work experiences and projects they are implementing in the “Computing Systems”.