Modern data science pipelines employ a variety of workloads, including tensor algebra, graph processing algorithms, and relational query processing. This results in using a set of loosely coupled data processing frameworks that move the data across the analytics pipeline, leading to unnecessary resource and energy consumption. This talk shows a compilation-based approach to move the computation closer to the data. This is achieved by designing (domain-specific) languages that leverage the structure of data with algebraic optimizations. We show that for a wide range of applications, including database query processing and tensor processing, our proposed approach significantly outperforms the state-of-the-art frameworks.
*Speaker Bio*: Amir Shaikhha is an Assistant Professor (Lecturer) in the School of Informatics at the University of Edinburgh. His research focuses on the design and implementation of data-analytics systems by using techniques from databases, programming languages, compilers, and machine learning communities. Prior to that, he was a Departmental Lecturer at Oxford. He earned his Ph.D. from EPFL in 2018, for which he was awarded a Google Ph.D. Fellowship in structured data analysis, as well as a Ph.D. thesis distinction award. He has won the Best Paper Award at GPCE 2017 and the Most Reproducible Paper Award at SIGMOD 2017. He (co-)chaired the program committees of DBPL 2021, Scala 2022, and GPCE 2023.
To join the video meeting, click this link: https://meet.google.com/oms-poez-bas
