Supercomputing Software for Moore and Beyond
Prof. Didem Unat
Parallel and Multicore Computing Laboratory
Koç University, Turkey
"Supercomputing Software for Moore and Beyond"
Wednesday, February 15, 2:00-2:40 pm UTC (30min talk + 10min questions)
6 am PST / 9 am EST / 2 pm UTC / 3 pm CET / 5 pm AST / 11 pm JST
Watch this webinar here and download the slide deck here!
Supercomputing Spotlights is a new webinar series featuring short presentations that highlight the impact and successes of high-performance computing (HPC) throughout our world. Presentations, emphasizing achievements and opportunities in HPC, are intended for the broad international community, especially students and newcomers to the field. Supercomputing Spotlights is an outreach initiative of SIAG/Supercomputing (https://siag-sc.org) … Join us!
Abstract: Thanks to the predictable trend of Moore’s Law during the past decades, supercomputers have shown great performance increases, opening the doors for incredible innovations and discoveries. However, due to unavoidable limits in chip manufacturing and rising energy consumption, the techno-economical model of Moore’s Law is (arguably) ending, stagnating the growth of computing and thus supercomputers. Regardless of the pessimism around Moore’s Law, as a community, we need to prepare our applications for what’s there and what’s upcoming. The arithmetic processing capabilities and complexity of systems continue to increase as multicore and accelerators are added to systems. Moreover, the improvement rates of both memory bandwidth and latency are much slower than that of the computing power, and the Post-Moore era makes data locality even more important. These trends urgently call for innovative software solutions to improve the usability and efficiency of supercomputers, and require doubling software efforts for more data locality-aware and seamless cooperation on both existing and emerging architectures. In this talk, I will explain the history of supercomputing in the context of data locality, our community efforts in the PADAL-series of workshops, and how we have shifted over the years from compute-centric to more data locality-centric programming. And yet, much more research is needed to push the sphere of supercomputing toward being more data-conscious.
Bio: Didem Unat is a faculty member at Koç University and director of the Parallel and Multicore Computing Laboratory. She is the first researcher from Turkey to receive ERC funding from the European Research Council in the field of Computer Science for her project BEYONDMOORE for 2021-2026. She is currently acting as project coordinator of the EuroHPC partnered project of 2.6M €. She was named the Emerging Woman Leader in Technical Computing by ACM SigHPC in 2021, the first recipient of this award outside the US. She is known for her work on programming models, performance tools, and system software for emerging parallel architectures. She received her PhD degree from the University of California, San Diego, and later the Luis Alvarez Postdoctoral Fellowship from LBNL. She received the Marie Sklodowska-Curie Individual Fellowship from the European Commission in 2015, the BAGEP Award from the Turkish Academy of Sciences in 2019, the British Royal Society Newton Advanced Fellowship in 2020, and the young scientist of the year award in Turkey in 2021.
