Philippos Papaphilippou

I am looking forward to supervise student projects with topics on FPGAs, sorting algorithms, network switches, multi-processor architectures and data science. With recent industrial knowledge on bleeding edge technology, I have recently joined TCD as an assistant professor to help with the Human Capital Initiative. I was a senior CPU architect at Huawei Technologies R&D (UK) Limited, and a contractor at Microsoft. I received my PhD from Imperial College London with funding from dunnhumby (Tesco) for researching novel accelerators to improve the performance of big data analytics.

Instruction usage analysis on today’s computers

Philippos Papaphilippou (email | all projects)

A general purpose processor (CPU) uses a pre-defined instruction-set architecture (ISA) to run common workloads with a reasonable performance. For certain workloads, this performance is CPU-bound, meaning that the pre-defined instructions are not able to use the silicon efficiently. In such cases, an accelerator including GPUs or FPGAs would be appropriate to improve the performance … Read more

Adaptive hardware algorithms for instruction-level parallelism in CPUs

Philippos Papaphilippou (email | all projects)

In this project the student will explore three types of adaptive mechanisms that exist in modern general purpose processors (CPUs): branch prediction (prediction of condition results to help with fetching near-future instructions) and cache replacement (evicting the cache blocks that are less likely to be referenced soon). The main task in this project is to … Read more

Improving GPU merge sort

Philippos Papaphilippou (email | all projects)

Sorting is a fundamental operation undeniably important for a variety of applications, including databases, security and experimental physics. This project is to develop or port a state-of-the-art algorithm to graphics processing units (GPUs). The idea is to accelerate sorting on numbers or key-value tuples while surpassing the performance and efficiency of existing algorithms. The goal … Read more