This PhD project at the University of Southampton, supervised by Dr Philippos Papaphilippou and Prof Steve Gunn, focuses on advancing instruction-level parallelism (ILP) in manycore processors. As CPUs with hundreds of cores become increasingly prevalent in applications ranging from embedded AI devices to large-scale servers, the need for efficient hardware prediction algorithms is critical. While current mechanisms such as data prefetchers and cache replacement policies have significantly improved single-thread performance, there remain substantial research gaps in optimizing ILP for manycore architectures. The project aims to develop novel prediction algorithms specifically tailored for manycore processors, addressing the theoretical limits and practical design attributes that maximize performance. Manycores offer high energy efficiency and programmability for parallel workloads, making them ideal for demanding applications in artificial intelligence and beyond. However, the rapid increase in computing capacity has outpaced advancements in main memory technology, necessitating innovative approaches to fully utilize available memory bandwidth and minimize latency. Through analytical modeling and hardware algorithm development, the research will establish new foundations for ILP in future manycore systems. The outcomes are expected to benefit highly-demanding workloads, ensuring that processors can execute more instructions per cycle and achieve optimal performance. Applicants should have a UK 2:1 honours degree or its international equivalent in computer science, computer engineering, electronic engineering, or a closely related field, with a strong emphasis on system and computer architecture. The School of Electronics and Computer Science is committed to equality, diversity, and inclusivity, welcoming applicants from all backgrounds and offering flexible working patterns and generous support policies. Funding opportunities include Horizon Europe fee waivers for qualifying students, competition-based Presidential Bursaries, and studentships covering UK-level tuition fees and living stipends. Funding is awarded on a rolling basis, so early application is encouraged. To apply, select the PhD Computer Science (7089) programme for 2026/27, Faculty of Engineering and Physical Sciences, and include the supervisor's name in your application. Required documents are a research proposal, CV, two academic references, degree transcripts and certificates, and proof of English language proficiency if applicable. For general queries, contact [email protected], or reach out to Dr Philippos Papaphilippou at [email protected] for project-specific questions. The University of Southampton is recognized for its commitment to sustainability and employee well-being, holding the Platinum EcoAward and offering a range of benefits to support work-life balance.

This PhD project at the University of Southampton, supervised by Dr Aristide Akem and Professor Steve Gunn, aims to redefine in-network computing for compute-native 6G networks. The research focuses on evolving network infrastructures from traditional packet-forwarding to distributed in-network computing platforms, enabling computation within the network data path and coordinating capabilities across edge and core networks. Over the past decade, significant progress has been made in offloading compute to programmable data-plane devices such as switches and SmartNICs. These advances have demonstrated the feasibility of embedding data aggregation, filtering, machine learning inference, and lightweight processing into packet-processing pipelines. However, challenges remain in scalability, coordination across distributed domains, predictability under strict latency constraints, and trustworthiness in multi-tenant environments. Addressing these challenges is essential for making in-network computing a native architectural feature of 6G networks. The project will systematically investigate the architectural and practical boundaries of in-network computing across mobile, edge, and core environments. It will characterise current constraints in data-path programmability, state management, synchronisation, and hardware capabilities, and design mechanisms to extend these capabilities while preserving performance guarantees, predictability, and verifiability. Central research questions include: identifying the fundamental scalability, programmability, and performance limits of current in-network computing platforms; synchronising and coordinating distributed in-network functions across edge and core domains; and developing abstractions and programming models for secure, trustworthy, and multi-tenant deployment of in-network computing functions. The research combines programmable networking, machine learning, distributed systems, and hands-on experimentation with modern networking hardware. Students will join a dynamic team with collaborations across the UK and abroad, aiming to publish in leading venues such as INFOCOM, NSDI, CoNEXT, SIGCOMM, and IEEE/ACM Transactions on Networking. Entry requirements include a UK 2:1 honours degree or its international equivalent, demonstrable programming experience in Python or C/C++, and prior coursework or project experience in computer networking or mobile networks. Desirable qualifications include experience with programmable networking technologies (P4, DPDK, eBPF, XDP), networking hardware (SmartNICs, switches, FPGAs), and machine learning frameworks. English language qualification is required if applicable. The University of Southampton offers a range of funding opportunities for both UK and international students. Horizon Europe fee waivers automatically cover the difference between overseas and UK fees for qualifying students. Competition-based Presidential Bursaries and studentships typically cover UK-level tuition fees and a stipend for living costs for top-ranked applicants. Funding is awarded on a rolling basis, so early application is recommended. To apply, select programme type (Research), 2026/27, Faculty of Engineering and Physical Sciences, search for PhD Computer Science (7089), and add the supervisor's name in section 2. Applications should include a CV, two academic references, degree transcripts and certificates, and English language qualification if applicable. For questions, contact the Faculty or project leader. The School of Electronics & Computer Science is committed to equality, diversity, and inclusivity, welcoming applicants from all backgrounds and offering flexible working patterns, generous maternity policy, onsite childcare, and a range of benefits to support well-being and work-life balance. The University is also committed to sustainability, holding the Platinum EcoAward.