Seminar: “Delay-Resistant Geo-Distributed Analytics”

CS Seminar: “Delay-Resistant Geo-Distributed Analytics” by Dr. Habib Mostafaei, Eindhoven University of Technology

Abstract: Big data analytics platforms have played a critical role in the unprecedented success of data-driven applications. However, real-time and streaming data applications, and recent legislation, e.g., GDPR in Europe, have posed constraints on exchanging and analyzing data, especially personal data, across geographic regions. To address such constraints data has to be processed and analyzed in-situ and aggregated results have to be exchanged among the different sites for further processing. This introduces additional network delays due to the geographic distribution of the sites and potentially affecting the performance of analytics platforms that are designed to operate in datacenters with low network delays. In this talk, I will present our recent work showing that the three most popular big data analytics systems (Apache Storm, Apache Spark, and Apache Flink) fail to tolerate round-trip times of more than 30 milliseconds even when the input data rate is low. We show that it is possible to improve the performance of all these popular big data analytics systems significantly amid even transcontinental delays (where inter-node delay is more than 30 milliseconds) and achieve performance comparable to this within a datacenter for the same load.

Bio: Habib Mostafaei is currently a tenured Assistant Professor of Computer Science at the Eindhoven University of Technology. He received the Ph.D. in Computer Science and Engineering from Roma Tre University in 2019. Before joining TU/e, he was a postdoctoral researcher at Technische Universitat Berlin, receiving a DFG grant to work on the BIFOLD-BBDC project from 2019-2022. He worked as a full-time faculty member at the Computer Engineering Department of Azad University from 2009-2015. He recently received a grant from Intel to develop high-speed networked systems. His main research fields include networked systems, network management, and distributed systems.

This Talk is part of the Computer Science department series of seminars.

This is an online Talk over Microsoft Teams meeting

Phone Conference ID: 998 180 264#