Research Interests

Improved accessibility to  heterogeneous platforms (CPUs, GPUs and FPGAs) via smarter tools and compilers.

Acceleration of Scientific Models and Machine Learning on Heterogeneous Platforms.

Exploring High-level Programming and Synthesis for FPGAs.

Performance and Cost Modeling.

Applied HPC  (atmospheric sciences, computational biology, quantum computing simulation, natural language research).

FPGA-based embedded systems.

Computing Science Education & Work-Based Learning

My main research interests are listed in the sidebar.

I am also engaged in a number of related research interests through collaborators and students:
    •  performance benchmarks for heterogeneous devices
    •  using FPGA clusters for HPC
    •  acceleration of atmospheric and oceanic models
    •  accelerating quantum computing simulation
    •  exploring optimized arithmetic for running scientific models on FPGAs
    •  accelerating neural networks on FPGAs
    •  type-aware System-on-Chip design
    •  secure e-voting systems
    •  developing FPGA based embedded systems for MEMs sensors
    •  using machine-learning for investigating bilingualism and bilingual grammar
If you are interested in doing a PhD in a related area, please get in touch.

The TyTra Project

From 2014-19, I was a post-doctorate researcher for the EPSRC-funded project, “Exploiting Parallelism through Type Transformations for Hybrid Manycore Systems”. The Principal Investigator for the project was Dr Wim Vanderbauwhede of the Department of Computing Science at the University of Glasgow. I am still continuing research in this area, following on from the project.

The project was in collaboration with Imperial College and Heriot-Watt University. The specific challenge that is addressed in this project is how to exploit the parallelism of a given computing platform, e.g. a multicore CPU, a graphics processor (GPU) or a Field-Programmable Gate Array (FPGA), in the best possible way, without having to change the original program. 

The project spans a number of interesting areas:
  • High performance computing (multi-core CPUs, GPGPUs, FPGAs)
  • Acceleration of scientific applications; current focus on atmospheric sciences
  • High level programming of FPGAs
  • Multi-party Session types, specifically their application to high-level program transformation for FPGA programming
  • OpenCL for multi-core, GPGPUs and FPGAs
  • Automatic transformation of high-level scientific code scientific applications
My work brings me in touch with all of these areas, but my focus is towards creating a cost-model based automated back-end for FPGAs. I am working on an architectural abstraction for the FPGA that is best positioned to exploit parallelism, especially in the context of scientific computations, and to make this architecture programmable in a way that it can be done automatically.

Research Groups

I am associated with the  Glasgow Parallelism research group (GPG) in the Glasgow Systems Section (GLASS) of the School of Computing Science.

Other Projects and Collaborations

Dr Wim Vandrebauwede has been my supervisor, line manager, and now one of my closest collaborators.

Along with Wim, I collaborate with Dr Rene Steijl from the School of Engineering at Glasgow, looking into the use of FPGAs for accelerating quantum computing similations, in the context of Computational Fluid Dynamics.

I was a Visiting Researcher at the Institute for Gravitational Research at University of Glasgow in the summer of 2018, providing FPGA related development support to Professor Giles Hammond and Dr Abhinav Prasad for developing a highly sensitive gravimeter for both ground and space based applications. The work was on developing a prototype solution on Microsemi's space-compatible SmartFusion2 SoC-FPGA.

I visited the NSF Center for High-Performance Reconfigurable Computing (CHREC) at the University of Florida in the summer of 2017, funded by a HiPEAC grant, working with Professor Herman Lam, with whom I remain connected.

I am collaborating with Professor Saji Hameed at the University of Aizu, Japan, along the lines of higrmance computing (HPC) for weather models. I have visited at University of Aizu as a visiting researcher, and then hosted him in Glasgow as a Royal Academy of Engineering  Distinguished Visitor.

I am a collaborator at  Dr Safee Ullah's Biomedical Informatics Research Laboratory (BIRL) at LUMS, Lahore, Pakistan. Our collaboration is in the area HPC for computational cancer biology,  and I am a project partner in the ICT  R&D funded project "Design and Development of a Next-Generation Modelling and Simulation Platform for Cancer Systems Biology".

I am actively collaborating with Dr Clara Cohen and Dr Catherine Higham on using machine learning to uncover interesting features of bilingual grammar. We have a couple of workshops papers together and are writing up a grant application.

I collaborate with Dr Syed Taha Ali in the Department of Electrical Engineering at NUST, Pakistan, where I support his work investigating secure e-voting hardware and software, in a Pakistani context.

I am a Glasgow University / ISLI alumnus and my Doctorate project was on creating a Dynamically Reconfigurable Platform for the MAC layer.

Publications

Available on my University profile page. Also: researchgate, Google Scholar profiles.

(Here's my doctorate (EngD) the thesis at Glasgow University's repository.)