University of Oxford
Oxford, UK
3.5-year D.Phil. studentship Supervisors: Prof Noa Zilberman The training of new AI models, as well as their deployment for inference, is transforming the design of computer networks. In this project, you will investigate innovative networking solutions for frontier AI systems, with a particular focus on the architecture of network switches. You will explore the role of the network switch in AI systems with sub-1 ns latency between accelerators, including how operations are partitioned between switches and edge devices. Building on this analysis, you will develop novel networking solutions that accelerate frontier AI systems. Eligibility This studentship is funded through the VMware University Research Fund and is open to Home and Overseas students (full award – fees plus stipend). Award Value Course fees are covered at the level set for Home or Overseas students. The stipend (tax-free maintenance grant) is the UKRI minimum stipend, and at least this amount for a further two and a...

