Ruhr University Bochum · Faculty of Computer Science
Data-Intensive Systems
Our vision
Build the next-generation of efficient and smarter data-intensive systems using artificial intelligence.
People
© Quang Duc Tran 
© Quang Duc Tran 
© Quang Duc Tran
Selected Publications
All 50 publicationsSamsara: Multimodal Stream Processing
Demonstrating Samsara: A Super-Optimizer for Multimodal Stream Processing
This paper received the Best Demo Award at VLDB 2026.
@article{ferri2026samsaradem, author = {Ferri, Alessandro and {dos Santos}, Uélison Jean Lopes and Nistor, Szilard and Agnihotri, Pratyush and Luthra, Manisha and Binnig, Carsten}, journal = {Proc. VLDB Endow.}, title = {Demonstrating Samsara: A Super-Optimizer for Multimodal Stream Processing}, volume = {19}, year = {2026}, url = {https://www.vldb.org/pvldb/vol19/p4666-santos.pdf} }Towards Multimodal Stream Processing Systems
@inproceedings{santos2026multimodalstream, author = {{dos Santos}, Uélison Jean Lopes and Ferri, Alessandro and Nistor, Szilard and Tommasini, Riccardo and Binnig, Carsten and Luthra, Manisha}, booktitle = {EDBT 2026}, title = {Towards Multimodal Stream Processing Systems}, year = {2026}, url = {https://www.openproceedings.org/2026/conf/edbt/paper-218.pdf} }
Learned Cost Models for Stream Processing Systems
COSTREAM: Learned Cost Models for Operator Placement in Edge-Cloud Environments
@inproceedings{heinrich2024costreamicde, author = {Heinrich, Roman and Binnig, Carsten and Kornmayer, Harald and Luthra, Manisha}, booktitle = {IEEE ICDE 2024}, title = {COSTREAM: Learned Cost Models for Operator Placement in Edge-Cloud Environments}, year = {2024}, url = {https://ieeexplore.ieee.org/document/10598052/} }ZeroTune: Learned Zero-Shot Cost Models for Parallelism Tuning in Stream Processing
@inproceedings{agnihotri2024zerotuneicde, author = {Agnihotri, Pratyush and Koldehofe, Boris and Stiegele, Paul and Heinrich, Roman and Binnig, Carsten and Luthra, Manisha}, booktitle = {IEEE ICDE 2024}, title = {{ZeroTune}: Learned Zero-Shot Cost Models for Parallelism Tuning in Stream Processing}, year = {2024}, doi = {10.1109/ICDE60146.2024.00163}, url = {https://ieeexplore.ieee.org/document/10598043/} }
Recent News
- October 2026People
Kota Hayashida visits us as an exchange student. Welcome!
Kota joins the Data-Intensive Systems Group as an exchange student researcher from the University of Tsukuba, Japan.
- September 2026People
Aashutosh Aryal joins the group. Welcome!
Aashutosh joins the Chair of Data-Intensive Systems as a student assistant.
- September 2026Award
Samsara, our multimodal super-optimizer demo, wins the Best Demo Award at VLDB 2026 🎉
Recognized among all demonstrations at VLDB 2026 for showing how Samsara optimizes multimodal stream processing queries in real time.
- August 2026Group
The group attended the VLDB Summer School
Members of the Data-Intensive Systems Group took part in the VLDB Summer School.
- August 2026People
Paul Adler joins the group. Welcome!
Paul joins the Chair of Data-Intensive Systems as a PhD student.
- June 2026People
Luca Witt-Schönfeld joins the group. Welcome!
Luca joins the Chair of Data-Intensive Systems as a PhD student.
- April 2026Appointment
Manisha Luthra Agnihotri appointed as Junior Professor at Ruhr University Bochum
The Chair of Data-Intensive Systems starts at the Faculty of Computer Science, building a research group on next-generation data systems that combine classical data system principles with modern AI.
Interested in working with us?
We are looking for PhD students and postdocs, host research interns, and supervise theses and student projects — currently on how AI agents can fundamentally change data systems.