Abstract: The next generation of multi-science hubs such as the SKA are based on a highly challenging operating principle, generating extreme scale data volumes to be processed and reduced in a time constrained manner, while remaining as energy efficient as possible and operable for decades. To build these infrastructures, the complete digital continuum has to be realized, through a hierarchy of cyber-infrastructures, fed continuously by science sensors at the edge; filtering, combining, processing and reducing continuous data streams in quasi real-time locally on supercomputers; and generating science grade data products to be delivered worldwide to a distributed community of scientists, relying on cloud infrastructures for high-level analysis. The challenges faced by international scientific communities in looking towards the implementation and exploitation of such experiments are driven by i) a paradigm shift to a dataflow model in which data streams have to be reduced into actionable data products before being discarded, ii) managing a new scale of data volumes, up to the multi-Exabyte as well as iii) fitting development, operations, maintenance and upgrades within a restricted cost and power envelope. In this talk, I will cover more quantitatively these challenges and will present the large scale R&D initiatives initiated at national and European levels to address them.

Bio: Damien holds a PhD in Observational Astronomy from Université Paris-Diderot (2005). As a technology enthusiast with extensive international experience, Damien has successfully spearheaded numerous disruptive R&D projects in astronomical instrumentation for almost two decades. His expertise lies in observational astronomy, high performance real-time cyber-physical systems, and artificial intelligence. He has acquired considerable experience in managing engineering teams at both micro and macro levels, coupled with a proven track record of training through research with more than a dozen successful PhD supervisions. Additionally, he has consistently secured multiple tenders from diverse funding sources at national and European levels, including multimillion-euro programs, public-private partnerships, and interdisciplinary R&D initiatives. Since 2021, with France officially joining SKAO, he is also getting strongly involved in the French effort dedicated to the construction of this giant radio-telescope. In particular, he is currently the director of ECLAT, a joint laboratory between CNRS, Inria, Observatoire de Paris, Observatoire de la Côte d’Azur and Atos/Eviden, as a long-term support structure federating resources from academic and industrial teams that will engage in the R&D work for the French contribution to the SKA.

Abstract: Distributed systems, such as Cloud computing, are increasingly spanning worldwide, with digital services hosted all around the globe and often belonging to complex systems, utilizing many other services and hardware resources themselves. Along with this increase comes an alarming growth of Cloud devices and their related energy consumption. Despite the distributed systems’ complexity, understanding how they consume energy is important in order to hunt wasted Joules and reduce their environmental impact. This talk will deal with measuring the energy consumption of distributed systems and deriving models from these measurements to evaluate their energy consumption and their environmental impact.

Bio: Anne-Cécile Orgerie is research scientist at CNRS, in the IRISA laboratory in Rennes. She got her PhD in 2011 in Lyon. She belongs to the Magellan team, dealing with large-scale distributed systems, Cloud computing and edge infrastructures. Her research interests include measuring, modeling, simulating and improving the energy efficiency of ICT distributed systems. She is currently the director of the working group EcoInfo studying the environmental impact of ICT devices.