The Role of an Autotuning Compiler in Getting to Exascale
Professor Mary Hall, School of Computing, University of Utah
As the cost of moving data on current and future architectures becomes increasingly dominant, the challenges of developing high-performance applications are increasingly onerous.
Future exascale architectures will look radically different from today’s platforms, and it is likely that different vendors will provide radically different hardware solutions (e.g., Intel Knights Landing vs. Nvidia GPUs).
The goal of compiler optimisation in high-performance computing is to take as input a computation that is architecture independent and maintainable and produce as output efficient implementations of the computation tuned for the target architecture.
Autotuning empirically evaluates a search space of possible implementations of a computation to identify the implementation that best meets its optimisation criteria (e.g., performance, power, or both).
Combining the two concepts, autotuning compilers generate this search space of highly-tuned implementations either automatically or with programmer guidance.
This talk will explore the role of compiler technology in achieving very high levels of performance, comparable to what is obtained manually by experts. As a case study, it will highlight some of the aggressive optimisations required to reduce communication for a specific high-performance application domain that is notoriously memory bound: geometric multigrid and the stencil computations within them. We will also highlight a path towards making this technology mainstream.
Mary Hall is a professor in the School of Computing at the University of Utah.
She received a PhD in Computer Science from Rice University.
Her research focuses on compiler technology for exploiting performance-enhancing features of a variety of computer architectures. Recent emphasis is on autotuning compiler technology for high-performance computing.
Professor Hall is an ACM Distinguished Scientist and an ACM representative on the Computing Research Association Board of Directors.
She is deeply interested in computing history, having served on the ACM History Committee since 2005 and as chair from 2009–2014. She also actively participates in outreach and mentoring programs to encourage the participation of women and under-represented minorities in computer science.