Skip to main content

Postgraduate research project

Efficient application mapping for massively parallel compute systems

Funding
Competition funded View fees and funding
Type of degree
Doctor of Philosophy
Entry requirements
2:1 honours degree View full entry requirements
Faculty graduate school
Faculty of Engineering and Physical Sciences
Closing date

About the project

In this research project, you'll develop effective mapping algorithms for graph-based problems. To achieve this, you'll work with multiple existing massively parallel hardware architectures at the University of Southampton.

Massive parallel compute architectures such as GPUs and FPGAs are becoming increasingly popular, as they offer greater parallelism capabilities and more application-specific processing capability, compared to traditional high-performance computing approaches.

Graph-based problems are particularly amenable to massive parallelism: a large number of small processing elements performing simple computation that interact with each other using small messages.

These include (and are certainly not limited to):

  • agent-based models
  • finite-element modelling
  • and cellular automata problems

Despite the ubiquity of graph-based massive parallel problems, determining how to map such problems onto hardware remains a largely unexplored problem, in part because hardware has only recently become available able to support the truly massive parallelism.

This is a major research opportunity in the development of efficient, general-purpose mapping algorithms for parallel systems that can start from an abstract graph representation of the problem and produce a hardware graph representing the problem as mapped to the physical hardware.

The goal will be to derive general principles for mapping, so that parallel programs can be designed using a structured approach, rather than through a collection of heuristics and ad-hoc techniques. These outcomes will be directly exploited and tuned on a portfolio of real problems outside computer engineering, encouraging collaboration and interdisciplinary research.

This knowledge is of immediate value in many industrial applications (including computational chemistry and machine learning): there is potential to reduce the compute times in the large-scale simulations from days to minutes, using much more economical hardware than the supercomputers in common use for such problems.

To perform this research, we require you to have completed an engineering degree in a relevant discipline, and to demonstrate an aptitude for computational modelling and software engineering. Knowledge of numerical optimisation techniques and place-and-route approaches are also desirable.

Working in an engineering research team should inspire and excite you. In return, we provide a team of computer engineering researchers who will provide expert, tailored guidance, a wealth of dedicated massively-parallel compute infrastructure, and industry links through external collaborators and the Arm-ECS research centre.

The Arm-ECS Research Centre was established in 2008, and is an award-winning research collaboration between the University and Arm.

The PhD student will be based primarily at the University of Southampton, UK, but will be supervised by researchers at both the University and at Arm, and offer the possibility of research placement(s) where appropriate.

Back
to top