03 Dec 2021, 13:30 Athens time, via Zoom
Resource-allocation Algorithms & Systems for Real-time Processing over Distributed Architectures
Many diverse applications from the life science, financial, maritime and other domains produce large volumes of data with speeds reaching GBs/min. In extreme-scale scenarios data arrive at a number of networked computer clusters or clouds with heterogeneous compute resources. In turn, each cluster hosts a variety of Big Data platforms for parallel processing purposes. Practitioners and business analysts need to execute global analytics workflows across these infrastructures and derive real-time answers to timely support critical decision-making procedures. Therefore, there is an emerging need for software technologies and architectures which will unify the execution of analytics over these settings, appropriately balancing the needs for (a) just use of the available resources while running various workflows, (b) maintaining the execution performance at high levels so that the real-time nature of the workflows is preserved. In this talk, we are going to discuss algorithms for efficient resource utilization and allocation over parallel and geo-distributed infrastructures with diverse Big Data technologies. We will present open-source software frameworks implementing these algorithms for any network of clusters/clouds and for prominent Big Data platforms. We will discuss how our techniques have been incorporated in state-of-the-art data science platforms and demonstrate their usefulness in real-world application examples.
About the speaker
Nikos Giatrakos is a a Postdoctoral Researcher at the School of Electrical and Computer Engineering of the Technical University of Crete (Greece) and an Adjunct Researcher at the Information Management Systems Institute of Athena Research Center (Greece). He received the BSc degree in Computer Science from the University of Piraeus (Greece) in 2006, the MSc degree in Information Systems from the Athens University of Economics and Business (Greece) in 2007, and the PhD degree in Computer Science from the University of Piraeus (Greece) in 2012. His research interests include Big Data Analytics, Distributed Data Streams, Approximate Query Processing, Data Synopses, Complex Event Processing and Data Management Issues in Internet of Things platforms. He has worked as the main, co-principal investigator or co-coordinator of EU and national projects focusing in these areas. He was a recipient of the Best Demo Award in ACM CIKM 2020.