|
|
| Anno Accademico |
2025/2026 |
| Corso di Studi |
Dott. - MI (1380) Ingegneria dell'Informazione / Information Technology |
| Anno di Corso |
1 |
| Codice Identificativo |
063928 |
| Denominazione Insegnamento |
DATACENTER NETWORKING IN THE AI ERA |
| Tipo Insegnamento |
MONODISCIPLINARE |
| Crediti Formativi Universitari (CFU) |
5.0 |
| Programma sintetico |
This course equips students with the knowledge to analyze and design Datacenter Networks (DCNs) by exploring key architectural solutions, from traditional Cloud-driven models to emerging AI-centric infrastructures. Students will learn fundamental switching theory, covering both circuit and packet switching, and examine how AI workloads impact intra-DC traffic patterns. The course also introduces distributed AI training, including Deep Neural Networks (DNNs), Large Language Models (LLMs), and Collective Communication Libraries (CCLs). Finally, students will understand the challenges of scaling AI training across multiple datacenters over Wide Area Networks (WANs), addressing network-aware optimizations for efficient distributed computing. |
| Settori Scientifico Disciplinari (SSD) |
|
Codice SSD
|
Descrizione SSD
|
CFU
|
|
ING-INF/03
|
TELECOMUNICAZIONI
|
5.0
|
|
|
Scaglione
|
Nome
|
Programma dettagliato
|
|
Da (compreso)
|
A (escluso)
|
|
A
|
ZZZZ
|
Ibrahimi Mëmëdhe, Maier Guido Alberto
|
|
|
