Machine Learning based Parallelisation and HW/SW co-Design of Heterogeneous multi-cores

4 PhD studentships are available to study Machine Learning based Parallelisation and HW/SW co-Design of for Heterogeneous multi-cores. The studentship will be held under thesupervision of Prof. Michael O'Boyle, within the Institute for Computing Systems Architecture, at the School of Informatics, University of Edinburgh, to begin in 2011, start date flexible.

Project

The project will look at a variety of projects in the areas of parallelisation and co-design where machine learning is a key technique to select the best optimisation or design. Parallelisation topics include the development of new techniques for mapping parallelism using machine learning; investigating dynamic compilation in the prescence of workload as well as and prototyping OpenCL (or similar) implementations where appropriate. HW/SW co-Design projects include design space exploration of the compiler/heterogeneous architecture co-space, compiler-directed selection of hardware confifgurations and dynamic hardware configuration based on runtime load. Typical topics include:
  • Mapping parallelism to multi-core/GPUs
  • Discovering and mapping parallelism to OpenCL
  • co-Design space exploration of compiler/architecture heterogeneous multicore architecture
  • Compilation for runtime adaptive hardware
  • Combined compiler/runtime offline/online adaptation to flexible hardware.
  • JIT compilation in the presence of dynamic hardware.
    • The project topics are however flexible and can change based on the applicants' interests. The CaRD group at Edinburgh is internationally leading in the use of machine learning for compiler and architecture co-design and optimisation - this will form the backbone to this project.

    Candidate Profile

    Suitable candidates will have a strong first degree in Computer Science and a strong interest in parallel programming, design space exploration, optimizing compilers or machine learning. The exact topic of the PhD is flexible depending on the candidate's interests. We are looking for the brightest minds to pursue research in a cutting-edge arena. The anticipated start date is Sept 2011 but this is flexible
  • All positions are fully funded

    • Applying for the Studentship

    Candidates are encouraged to contact Michael O'Boyle to informally discuss the project further. Formal application will be through the School's normal PhD application process.