Open Lecture 7 – In-Network Computing and Intelligent User Plane for 6G

Name: Open Lecture 7 - In-Network Computing and Intelligent User Plane for 6G
Start: 2024-03-21
End: 2024-03-21
Location: Virtual event

Following the success of the previous lectures, the Association is happy to confirm that Open Lecture 7 will take place on March 21, 2024, at 14h00 CET, covering the subject of In-Network Computing and Intelligent User Plane for 6G.

After mobile edge computing, the next evolutionary step on the pathway to bring computation even closer to the user is to compute in the network. In-Network Computing (INC) is a promising enabler for 6G with high potential to support meeting the stringent delay requirements of 6G use-cases and applications, including robotics, VR gaming, or remote surgery. Besides latency reduction, the concept of INC has been proven beneficial for network traffic volume reduction and offloading compute tasks from mobile end-devices to the network. This allows to minimize the devices’ battery drainage and supports the design of more lightweight devices.

The one6G Open Lecture 7 will center around the topic of In-Network Computing and Intelligent User Plane for 6G. Four top experts in the field, representing both academia and industry, will discuss the state of the art of INC, including architecture concepts, current challenges, and recent advances when it comes to integrating native computing as a key feature into 6G networks.

Join us on March 21, 2024 at 2 p.m. CET to learn more!

AGENDA

14:00 Welcome & opening

Riccardo Guerzoni

Director of the Core Network Group, DOCOMO Euro-Labs, Munich

Riccardo Guerzoni holds an M.Sc. degree in Telecommunications Engineering from the Università degli Studi di Ferrara, Italy. Currently, he is the director of the Core Network Group at DOCOMO Euro-Labs in Munich. Riccardo contributed to the standardization of the 3GPP 5G core network since Release 15, with special focus on Edge Computing and Industrial IoT. Since June 2020, he has been a board member of the 5G Alliance for Connected Industries and Automation (5G-ACIA). He is currently active in different standardization research initiatives on next generation mobile networks, including one6G.

14:05 Native artificial intelligence and native computing as key features of 6G

6th Generation Cellular Networks will be designed as an evolution of the current 5th Generation. Still, improving existing features alone will be insufficient to trigger massive investment and mass deployment. While new target use cases and applications are extensively studied by the academic and industrial communities, the present paper addresses the question which novel and innovative key features will be offered to customers, dramatically changing the overall user experience. It is outlined how Artificial Intelligence (AI) and Distributed Compute features will be natively integrated into the network, providing immediate (low-latency) access to almost infinite resources. We further comment on a future AI-as-a-Service framework which enables users to request trained AI models explicitly tailored to their current context and need. Data owners are thus able to monetize their knowledge through entirely new services without exposing the underlying data itself.

Markus Mueck

Intel

Markus Mueck received the Dipl.-Ing. and ing. dipl. degrees from the University of Stuttgart, Germany and Ecole Nationale Supérieure des Télécommunications (ENST), Paris, France in 1999. In 2006, he received the Doctorate degree of ENST in Communications. He is a Principal Engineer with Intel Deutschland, Munich, Germany and an Engineering Director; he acts as the Chair of the ETSI Board, he is a Chair of ETSI OCG AI coordinating the overall effort of the organization on AI and Adj. Professor of University of Technology, Sydney, Australia. Markus was leading the effort on Artificial Intelligence and Machine Learning of the European funded research Flagship Project Hexa-X preparing the design of 6th Generation Mobile Networks (6G).

14:25 Intelligent programmable data planes to optimize application delivery in 6G networks

Ever since the delivery of the first fully programmable network switch and the development of the P4 language “computing in the network” has transformed network operations and management. Now adding intelligence to the dataplanes represents a new advance in networking architecture, leveraging the programmability, Machine Learning (ML) as well as telemetry to embed intelligence and data-driven decision making directly into network devices. In traditional networking, dataplanes primarily focus on forwarding packets based on predetermined rules. However, intelligent dataplanes augment this functionality by enabling dynamic decision-making in the core-edge continuum. They can analyze network traffic patterns, predict future demand, detect anomalies, and optimize traffic routing in real-time. The talk will introduce Intelligent Data Planes and give examples and use cases from emerging applications in Internet of Things (IoT), edge computing, and 6G networks. Intelligent Data Planes have the potential to greatly improve network performance by making networks more responsive, efficient, and adaptive to application requirements, thus delivering the promise of joint network/application optimization.

Marie-Jose Montpetit

McGill University, Montreal, Canada

Marie-José Montpetit is a well-known Canadian researcher in distributed systems, in-network computing, digital twinning and applied artificial intelligence. From 2009 to 2019, she was a research scientist at MIT, and before that, she held numerous R&D positions in the wireless networking industry. She now focuses her research on distributed intelligent networking and applications. She is the co-chair of the Computing in the Network (COIN) research group at the IRTF, an adjunct professor at École de Technologie Supérieure in Montreal, a lecturer at Telecom Paris Sud on programmable networks at and McGill University on the mapping of networking intelligence to supply and distribution and an academic consultant at Iowa State university on 6G intelligent wireless networks. She is the recipient of the 2008 Motorola Innovation Prize and the 2010 MIT TR10.

14:45 In-network service-aware computing with hardware acceleration in 6G networks

Ubiquitous computing is a common and symbolic application included in the 6G framework. The computing function in the 6G network will be the enabler for many types of services in the 6G era, such as cyber-physical systems or AI-integrated immersive communications. The network computing platform applying to RAN will finally be shared with application functions and not only with networking functions, and this will result in having tighter linkage of network and application functions in a common computing platform. As we move toward such a 6G environment, NTT proposed an “Inclusive Core” that integrates and coordinates computing and networking functions for a much better user experience of services in the 6G era by applying in-network computing services exploiting computing resources in the 6G network. The architecture features serverless orchestration of in-network computing service and a multi-hardware chaining acceleration using types of accelerators, such as GPU, DPU IPU, etc.

Hiroki Baba

Network Service System Laboratories, NTT

Hiroki Baba is a senior research engineer at Network Service System Laboratories in NTT. He has been engaged in mobile network architecture for more than 10 years. He also engaged in the design and planning of NFV commercial introduction at DOCOMO. He actively worked on network slicing-related projects in the MEF forum and ETSI ZSM. Currently, he is working on 6G mobile network architecture, especially on the theme of in-network computing. He received B.E. and M.E. degrees from the Tokyo Institute of Technology in 2006 and 2008, respectively.

15:05 Intelligent user plane for the manufacturing sector

In this presentation intelligent user plane and in network computing tailored for the manufacturing sector will be analysed and discussed. Industrial networks have several very peculiar characteristics and requirements and may be a significant vertical application for future networks and 6G. The Operation Technology networks still leverage many legacy protocols and devices and are very challeging because of the very strict continous operation constraint. The presentation will discusse a few use cases we developed with 5G, referring to remote maintainance and provisioning of ad hoc secure connectivity. By analysing these results the presentation will identify current limitations and open opportunities to be considered for the development of 6G.

Franco Callegati

University of Bologna

Franco Callegati is Full Professor of Computer Science at the University of Bologna. His research interest are in the field of Software Defined Networking (SDN), Network Function Virtualization (NFV) and cybersecurity. He has been active in EU-funded research projects since 1994, often coordinating working groups and work packages. He served as co-chair of the Education Committee of the IEEE SDN Initiative. He is active in technology transfer project and was Director of the Center for Industrial Research in ICT of the University of Bologna from 2013 to 2019.

15:25 Q&A session and closing remarks

Date

21 Mar 2024

Expired!

Time

2:00 pm

More Info

Registration

Location

Virtual event