I did my undergraduate degree (BSc) in Computer Science at Durham University, specialising in theoretical computer science. I wrote my dissertation, available here on evaluating the "raw" (i.e. unoptimised) speedup achievable by parallelising multi-agent systems on GPUs.
I am a student in the EPSRC Centre for Doctoral Training in Pervasive Parallelism, where I am currently in the masters phase, supervised by Christophe Dubach and Murray Cole. I am a member of ICSA and CaRD.
I'm interested in the intersection between high performance parallel code and programming language theory. Specifically, I'm interested at looking at how we can compile high level functional or skeletal languages to as efficient code as possible. For parallelism my main interest is in generating efficent code for GPU execution. Programming for the GPU presents a very different programming challenge compared to traditional parallel CPU programming, and approaching this from a high level is my primary interest.
For my Masters project I'm looking at how we can implement specialised sparse linear algebra on GPUs for the efficient execution of high level graph algorithms.
At the prompting of a friend, I've written out a version of this using only the 1000 most common english words.
Harries, A, Steuwer, M, Cole, M, Gray, A, Dubach, C 2016, ‘Compositional Compilation for Sparse, Irregular Data Parallelism’. in Workshop on High-Level Programming for Heterogeneous and Hi- erarchical Parallel Systems (HLPGPU) 2016 @ HiPEAC, Prague, Czech Republic, January 19, 2016
Address: Informatics Forum, Room 1.07, 10 Crichton Street, Edinburgh, United Kingdom. EH8 9AB