This paper explores how Telco networks can enable distributed Artificial Intelligence (AI) by equipping Central Offices with GPUs and leveraging network acceleration. We discuss Remote Direct Memory Access (RDMA) over Ethernet and its evolution towards Ultra-Ethernet, Data Processing Units (DPU), and in-network computing as key enablers to transform geographically distributed infrastructure into a cost-effective programmable AI platform.
Network Acceleration for AI Workloads in Telco Networks
Cugini F.;Paolucci F.;Degl'Innocenti A.;Ismail L.;Paolini E.;Sgambelluri A.;Castoldi P.;
2026-01-01
Abstract
This paper explores how Telco networks can enable distributed Artificial Intelligence (AI) by equipping Central Offices with GPUs and leveraging network acceleration. We discuss Remote Direct Memory Access (RDMA) over Ethernet and its evolution towards Ultra-Ethernet, Data Processing Units (DPU), and in-network computing as key enablers to transform geographically distributed infrastructure into a cost-effective programmable AI platform.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
Network_Acceleration_for_AI_Workloads_in_Telco_Networks.pdf
solo utenti autorizzati
Tipologia:
PDF Editoriale
Licenza:
Copyright dell'editore
Dimensione
1.4 MB
Formato
Adobe PDF
|
1.4 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
