Friday, 15th November 2013, 3:30 pm (SCE Grid Conference Room (N4-01C-10))
Speaker: Professor Georgios Theodoropoulos (Durham University, UK)
Title: Simulation in the Era of Exascale Computing and Big Data
Advances in VLSI and the emergence of powerful processing elements at a scale never available before, have brought a revolution to High Performance Computing. Many-core processors, GPUs and FPGAs have made Petascale computing a reality and have brought Exascale computing in the horizon with the first Exascale systems expected by the end of the decade. At the same time there has been an explosion in technologies for data generation, storage, processing and analysis at a very large scale crossing all known boundaries. The emergence of a new class of large scale data-intensive analytics applications coupled with the scalability and power requirements of the next generation computational platforms introduce new technological and methodological challenges. The talk will examine the role of simulation in the context of these emerging technologies and analytics approaches and will outline the challenges ahead.
Georgios Theodoropoulos is a Professor in the School of Engineering and Computing Sciences at Durham University, UK and an adjunct Professor in the School of Computer Science and Statistics at Trinity College Dublin, Ireland. At Durham University he is the Head of the Innovative Computing Group and the Executive Director of the Institute of Advanced Research Computing (iARC), a major initiative to develop cutting-edge inter-disciplinary research at the interface between computing and the physical and social sciences, the arts and the humanities. Prior to his current position, he was a Senior Research Scientist with IBM Research, playing a leading role in the company’s Exascale Systems Research programme. In the past he was a senior academic at the University of Birmingham, UK, where he set up and led the Distributed Systems Lab and the Midlands e-Science Center. In 2009 he spent a sabbatical year at NTU. His current research focuses on modelling, simulation and analysis of complex socio-technical systems of systems and co-design for workload optimised systems.