About the project
6G networks will be compute-native, evolving from packet-forwarding infrastructures into distributed in-network computing platforms. This project seeks to extend the limits of in-network computing by redefining what computation can be performed within the network data path, and how such capabilities can be coordinated across edge and core networks.
In-network computing enables performing computation within the network infrastructure. Over the past decade, substantial progress has been made towards offloading compute to programmable data-plane devices such as switches and SmartNICs, demonstrating that data aggregation, filtering, machine learning inference, and lightweight processing can be embedded into packet-processing pipelines.
However, important challenges remain. Scalability, coordination across distributed domains, predictability under strict latency constraints, and trustworthiness in multi-tenant environments must be addressed before in-network computing can become a native architectural feature of 6G networks. A rigorous understanding of these limits, and principled methods for overcoming them, remains a critical gap.
This PhD project will systematically investigate the architectural and practical boundaries of in-network computing across mobile, edge, and core environments, and develop principled approaches to overcome them.
The project will characterise current constraints in data-path programmability, state management, synchronisation, and hardware capabilities, and design mechanisms that extend these capabilities while preserving performance guarantees, predictability, and verifiability.
The research will address three central questions:
- what are the fundamental scalability, programmability, and performance limits of current in-network computing platforms, and how can they be extended to support native integration into next-generation network architectures?
- how can distributed in-network functions be synchronised and coordinated across edge and core domains while maintaining predictable and bounded behaviour?
- what abstractions and programming models are required to enable secure, trustworthy, and multi-tenant deployment of in-network computing functions?
The project combines programmable networking, machine learning, distributed systems, and hands-on experimentation with modern networking hardware to help make in-network computing native to 6G and beyond.
You will join a dynamic team with collaborations across the UK and abroad, and aim to publish research outputs in leading venues such as INFOCOM, NSDI, CoNEXT, SIGCOMM, and IEEE/ACM Transactions on Networking.
The School of Electronics & Computer Science is committed to promoting equality, diversity inclusivity as demonstrated by our Athena SWAN award. We welcome all applicants regardless of their gender, ethnicity, disability, sexual orientation or age, and will give full consideration to applicants seeking flexible working patterns and those who have taken a career break. The University has a generous maternity policy, onsite childcare facilities, and offers a range of benefits to help ensure employees’ well-being and work-life balance. The University of Southampton is committed to sustainability and has been awarded the Platinum EcoAward.
