Identifying And Overcoming Challenges In The Data Processing Landscape
System challenges in large-scale data analytics
John Liagouris, Research Scientist, Hariri Institute of Computing
Thursday, February 20, 2020
12:00pm – 1:00pm
Hariri Institute of Computing, Seminar Room, Room 180
Abstract: Data processing systems play a key role in democratizing data science. They provide users with the necessary abstractions to construct complex data pipelines, scale-out computations to large clusters, and extract meaningful insights from data via simple programming models. As these systems become more and more popular, users demand better performance, scalability, and reliability, but also for new functionality, such as output explanation, what-if analytics, interactive debugging, and online profiling.
In this talk, I will give an overview of my research efforts to address some of the challenges in the evolving landscape of data processing systems. I will focus on techniques that help users understand the semantics and performance of distributed computations, discuss early results in this direction, and conclude with ideas for future work.
Bio: John Liagouris is a research scientist at the Hariri Institute for Computing and an adjunct assistant professor at Boston University. His research interests lie in distributed systems and databases, with a particular focus on real-time analytics, explainability, performance analysis, and optimization. Before joining BU, he was a visiting scholar at the RISELab, UC Berkeley, a senior researcher (oberassistent) at the Systems Group, ETH Zurich, a visiting research fellow at the University of Hong Kong, and a research assistant at the Information Management Systems Institute (IMSI) of the Research Center ‘Athena’, Greece. John obtained a diploma in Electrical and Computer Engineering in 2008, and a Ph.D. in 2015, both from NTU Athens, Greece.