Don’t Settle for Less, Fully Exploit Emerging Hardware with New System Software
Antonio Barbalace, University of Edinburgh
Today, computer architectures are drifting away from the von Neumann model. Computers, from mobile devices to servers in the data-center, are increasingly built with a multitude of different general-purpose, special-purpose, and even reconfigurable processing units – which share memory, have their own memory, or are tightly integrated with memory, such as near data processors. Moreover, IO is becoming faster and faster. This hardware (r)evolution, which is the response of computer architects to the growing applications’ demands, clearly affects fundamental assumptions on which the software we are using today is built upon.

With changing computer hardware, system software (hypervisor, operating system, compiler, and runtime library) can play an essential role to enable application software to fully exploit emerging computer architectures, while minimizing application complexity. Thus, potentially improving not only performance, energy efficiency, resource utilization, security, and fault-tolerance, but also programmability and portability. However, system software did not change much since decades.

This talk walks through Antonio Barbalace’s previous and current research on operating systems and compilers presenting how new system software designs can improve the scalability of parallel hardware, the programmability of heterogeneous compute units, and the performance and energy efficiency of near data processing platform – while reducing applications’ complexity.

Please email for a Zoom link
About the speaker
Antonio Barbalace is a Senior Lecturer (Associate Professor) at the School of Informatics of The University of Edinburgh, Scotland. Before, he was an Assistant Professor in the Computer Science Department, at Stevens Institute of Technology, New Jersey. Prior to that, he was a Principal Research Scientist and Manager at Huawei’s Munich Research Center (MRC). He was a Research Assistant Professor, and before a Postdoc, within the Systems Software Research Group, ECE Department, at Virginia Tech, Virginia. He earned a PhD in Industrial Engineering from the University of Padova, Italy, and MS and BS in Computer Engineering from the same University.

Antonio Barbalace’s research interests include all aspects of system software, embracing hypervisors, operating systems, runtime systems, and compilers/linkers, for emerging highly-parallel and heterogeneous computer architectures, including Near-Data Processing (NDP) platforms. His research seeks answers about how to architect or re-architect the entire software stack to ease programmability, portability, enable improved performance and energy efficiency, determinism, fault tolerance, and security. His research work appeared at top systems venues including ASPLOS, EuroSys, VEE, ICDCS, Middleware, EMSOFT, HotOS, HotPower, and Ottawa Linux Symposium. At the University of Edinburgh, he is investigating new systems software to handle heterogeneous processing units in a single computers or in data-centers focusing on the software architecture, programming models and interfaces, and scheduling and mapping of workloads.
Date & Time
Thursday, March 25, 2021 - 14:00