I am a research associate in the CAPcelerate project at the University of Cambridge.

I am looking at how we can incorportate CHERI capabilities into future accelerators, as well as how CPUs integrating CHERI can mix with accelerators without support for capabilities.

Before my current position, I completed my PhD at the Institute for Computing Systems Architecture (ICSA) at the University of Edinburgh. I investigated pattern based parallel programming models (also known as algorithmic skeletons) and their benefits for

• multiprogrammed heterogeneous systems.
• parallel hard real-time system.

Publications

• Device Hopping: Transparent Mid-Kernel Runtime Switching for Heterogeneous Systems
  Paul Metzger, Volker Seeker, Christian Fensch, Murray Cole
  In Transaction on Architecture and Code Optimization (TACO), 2021. Paper
• Enforcing Deadlines for Skeleton-based Parallel Programming
  Paul Metzger, Murray Cole, Christian Fensch, Marco Aldinucci, Enrico Bini
  In Proc. of RTAS 2020 (Top tier real-time systems conference. Acceptance rate: 26.9%) Paper
• NUMA Optimizations for Algorithmic Skeletons
  Paul Metzger, Murray Cole, Christian Fensch
  In Proc. of Euro-Par 2018

Publications in other Fields

• Lifelogging for 'Observer' View Memories: An Infrastructure Approach
  Sarah Clinch, Paul Metzger, Nigel Davies
  In Adjunct Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing
• Collecting Shared Experiences through Lifelogging: Lessons Learned
  Sarah Clinch, Nigel Davies, Mateusz Mikusz, Paul Metzger, Marc Langheinrich, Albrecht Schmidt, Geoff Ward
  In Volume 15 of IEEE Pervasive Computing

Posters

• Plastic Algorithmic Skeletons
  Presented at IPDPS 2017 and the HiPeac ACACES summer school in 2017:
  Poster, Poster abstract

Awards