Researchers in the Database Systems and Information Management (DIMA) Group at TU Berlin presented two full research papers as well as a demo and two workshop papers at VLDB 2022 (https://vldb.org/2022/), the 48th International Conference on Very Large Data Bases, which took place in Sydney, Australia from September 5th to the 9th.
(Pictured from left to right: Moritz Ruge, Lennart Behme, Philipp Grulich, Mahdi Esmailoghli [now at the Leibniz Universität Hannover], Haralampos Gavriilidis, and Kajetan Maliszewski.)
This year DIMA researchers contributed two full research papers, a demonstration, and two workshop papers to VLDB, a CORE A* conference (i.e., one of the most prestigious conferences in the database sector).
The paper “What is the Price for Joining Securely? Benchmarking Equi-Joins in Trusted Execution Environments” authored by Kajetan Maliszewski, Jorge-Arnulfo Quiané-Ruiz, Jonas Traub, and Volker Markl benchmarks different join algorithms on secure hardware. In the paper, the authors discuss how to make Trusted Execution Environments practical for relational workloads in the context of join operators.
The paper “Babelfish: Efficient Execution of Polyglot Queries” authored by Philipp M. Grulich, Steffen Zeuch, and Volker Markl introduces an efficient execution engine for polyglot queries that combines relational operators and UDFs in different query languages. In particular, Babelfish performs holistic optimization across the boundary of individual operators to improve query efficiency. Consequently, Babelfish eliminates performance bottlenecks attributed to UDFs and achieves high performance.
Sergey Redyuk, Zoi Kaoudi, Sebastian Schelter, and Volker Markl offered a demonstration of DORIAN in action. DORIAN is a human-in-the-loop approach for recommending end-to-end data science pipelines to novice users and domain experts code-free. It addresses practical challenges, such as providing recommendations efficiently in real-time,
At CDMS, the 1st International Workshop on Composable Data Management Systems, Haralampos Gavriilidis and Lennart Behme presented their paper “Towards a Modular Data Management System Framework.” In the paper, they describe a new modular database system architecture, where individual components are decoupled. This new architecture enables the composition of new systems without having to reimplement all components from scratch, and hence both fosters reusability and facilitates the development of novel applications. The presented work was a result of an industry collaboration with Huawei.
At VLIoT, the 6th International Workshop on Very Large Internet of Things, Ariane Ziehn, Christian Mandel, Kathrin Stich, Rolf Dembinski, Karin Hochbaum, Steffen Zeuch, and Volker Markl presented their paper “IoT-PMA: Patient Health Monitoring in Medical IoT Ecosystems.” IoT-PMA is an end-to-end solution that addresses the challenges of medical IoT applications by providing unified and ad-hoc views on a patient's vital parameters, including enhanced analysis and predictions to support timely physician decision-making. Furthermore, the authors envision the infrastructure of an IoT ecosystem for a multi-hospital scenario that enables geo-distributed medical participants to contribute data to the application in a secure, private, and timely manner.
Die Publikationen im Einzelnen: