Tuning parallel applications on software-defined supercomputers

Project opportunity

This Earmarked Scholarship project is aligned with a recently awarded Category 1 research grant. It offers you the opportunity to work with leading researchers and contribute to large projects of national significance.

This project aims to design and implement novel techniques and tools that allow programmers to understand and analyse their codes, with the goal of efficiently finding and correcting performance bottlenecks. Specifically, we will focus on software defined supercomputers in which performance degradation occurs as a result of memory virtualisation. Our goal is to make it easier to develop and optimise scientific codes on this class of machine, improving programmer productivity and therefore increasing the number of programs than can exploit such machines.

The overall ARC funded Linkage project aims to:

  1. Devise a novel framework that allows application developers to measure and analyse the effect of memory virtualisation on the performance of an application;
  2. Explore innovative tuning strategies that increase data locality and optimise data movement on virtualised node architectures;
  3. Design and implement powerful software tools that simplify the identification and location of data accesses that cause performance degradation; 
  4. Perform research on innovative methods that provide expert advice on how to optimise codes, and even optimise some applications automatically; and
  5. Demonstrate the effectiveness of the new approaches on applications in life sciences and aerospace

There is sufficient flexibility in the project to accommodate a PhD student working in one or more of these steps. At this stage the most likely options are in producing a hypervisor that virtualises distributed memory, or in designing novel performance visualisations and implementing them in a set of tools.

Scholarship value

As a scholarship recipient, you'll receive: 

  • living stipend of $32,192 per annum tax free (2023 rate), indexed annually
  • tuition fees covered
  • single Overseas Student Health Cover (OSHC)


Professor David Abramson

School of Information Technology and Electrical Engineering

Email: david.abramson@uq.edu.au

Preferred educational background

Your application will be assessed on a competitive basis.

We take into account your

  • previous academic record
  • publication record
  • honours and awards
  • employment history.

A working knowledge of computer science, including a detailed knowledge of some or all of supercomputing architectures, parallel programming, operating systems and scientific visualisation would be of benefit to someone working on this project.

The applicant will demonstrate academic achievement in the field(s) of computing systems and the potential for scholastic success.

A background or knowledge of computing architecture and memory subsystems is highly desirable.

Latest commencement date

If you are the successful candidate, you must commence by Research Quarter 2, 2023. You should apply at least 3 months prior to the research quarter commencement date.

If you are an international applicant, you may need to apply much earlier for visa requirements.

How to apply

You apply for this project as part of your PhD program application.

View application process