Abstract:

Modern data analytics and data science are at the heart of enterprise computing and drives advances in many scientific disciplines. In addition to designing new models and techniques for analyzing the data deluge, we also need system support that makes the analysis more efficient when executed on modern and future hardware.

Addressing such a challenge requires an effort that is beyond what is typically done within a single layer of the system stack. Applications are typically unaware of a machine’s runtime state, and the OS does not know what the goals of the applications are.

In her talk, Giceva will give an overview of her ongoing research on "Database/Operating System Co-Design" and show the benefits of a holistic approach by opening the interfaces and customizing the system stack for modern data processing workloads.

Bio:

Jana Giceva is a Ph.D. student in the Systems Group at Swiss Federal Institute of Technology in Zurich (ETH Zurich). Her research interests are in systems support for data science and big data to enable efficient use of modern and future hardware. The scope of her research spans multiple areas from the data processing layer to operating systems, which includes hardware accelerators for data processing.

In 2014, she earned a Google European Ph.D. Fellowship in operating systems.