Building 53, room 4025
Speaker: Dr Philippos Papaphilippou, University of Southampton, UK
Title: Big-data processing with future RISC-V processors and FPGAs
Abstract: The amount of data produced by humans per year is already exceeding hundreds of zettabytes (zetta=10^21), and a considerable percentage is processed for new information. This exponential growth has already changed the way we build and program computers, as the traditional computer architecture has been proven incapable of keeping up.
In this talk we will overview state-of-the-art parallel algorithms that help with big data analytics, especially with regard to sorting as a prominent use case. We will also see how these algorithms and methods are used by accelerators such as with Field-Programmable Gate-Arrays (FPGAs), as well as by forthcoming RISC-V processors.
Bio: Philippos has joined the University of Southampton as a lecturer in 2025. He received his PhD from Imperial College in 2021 and master's degree from the University of Cambridge in 2017. His doctoral studies were funded by dunnhumby (Tesco) for researching novel accelerators to improve the performance of customer analytics. After a contractorship at Microsoft and a two-year industrial position at Huawei as a CPU Architect, he has joined Trinity College Dublin as a temporary Assistant Professor in 2023.
His current research includes reconfigurable architectures, such as with FPGAs and their applications on data science, computer architecture and beyond.