Peter Pietzuch

What can machine learning systems learn from data analytics?

Machine learning models are becoming an integral part of data analytics pipelines, yet current machine learning systems are designed differently from established data management systems. They often have immature abstractions, which causes problems that data management systems have solved decades ago. In this talk, I will focus on two open challenges that machine learning systems face, namely elasticity and adaptability. 

I will describe our work that introduces new abstractions in machine learning stacks, heavily inspired by database technology, to address these problems, while remaining compatible with current machine learning platforms.

back to overview

Watch Recording



Peter Pietzuch is a Professor of Distributed Systems at Imperial College London, where he leads the Large-scale Data & Systems (LSDS) group ( His research work focuses on the design and engineering of scalable, reliable and secure data-intensive systems, with a particular interest in performance, data management and security issues. He has published papers in premier scientific venues, including OSDI, SOSP, SIGMOD, VLDB, USENIX ATC and EuroSys. He serves as the Director of Research in the Department of Computing and a Co-Director for Imperial's I-X initiative, and he was the former Chair of the ACM SIGOPS European Chapter (EuroSys). Before joining Imperial, he was a post-doctoral Fellow at Harvard University. He holds PhD and MA degrees from the University of Cambridge.