iiinova — where your digital self keeps evolving

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 28/04/2025

Open PhD position in the Systems Group! Focus is on using FPGAs (e.g., in Smart SSDs or Smart NICs) to address the divide between data growth and CPU stagnation as part of the recently launched CHORYS EU project. Apply now; deadline is 13.05.2025! www.career.tu-darmstadt.de/HPv3.Jobs/TU...?

career.tu-darmstadt.de

Wiss. Mitarbeiter:in (w/m/d) - Data Processing on FPGAs

Die TU Darmstadt steht für exzellente und relevante Wissenschaft. Die tiefgreifenden globalen Veränderungsprozesse – von Energiewende bis zu Künstlicher Intelligenz – gestalten wir durch herausragende...

042

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 10/03/2025

Paper accepted to the Datenbank-Spektrum: "Towards High-performance and Trusted Cloud DBMSs" We explore how modern Trusted Execution Environments (TEEs) like Intel SGX can be integrated into cloud DBMSs to enhance security while maintaining high performance. rdcu.be/ecEOa

020

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 07/03/2025

Systems Group @ BTW 2025 in Bamberg! We had a wonderful time, huge thanks to the organizers and speakers!

030

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 27/02/2025

Systems Group @ ELLIS RLGMSD workshop in Amsterdam to highlight how data engineering in enterprises is harder than public benchmarks make it look. Thanks to the organizers and speakers for the great event! Check out our papers: vldb.org/workshops/20... openreview.net/pdf?id=m85fY...

060

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 06/02/2025

Paper accepted to VLDB'25: "ELEET: Efficient Learned Query Execution over Text and Tables" ELEET is a query execution engine that integrates text sources into databases and achieves up to 575× speedups over LLM-based approaches without sacrificing accuracy. www.vldb.org/pvldb/vol17/...

040

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 04/02/2025

Paper accepted to SIGMOD'25: "How Good Are Learned Cost Models, Really? Insights from Query Optimization Tasks" Learned Cost Models often predict execution costs more accurately than traditional cost models. But does this translate into better query plans? Find out here: arxiv.org/abs/2502.01229

030

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 24/01/2025

We are proud to share that Johannes Wehrstein and Carsten Binnig received the CIDR'25 Best Paper Award for their paper "Towards Foundation Database Models", which is the result of joint work with SystemsResearch@Google! 🏆 @jwehrstein.bsky.social www.vldb.org/cidrdb/paper...

010

Systems Group @ TU Darmstadt @tuda-systems.bsky.social · 24/01/2025

Hello World! The Systems Group at TU Darmstadt is a collaboration of professors Carsten Binnig and Zsolt István from the Computer Science department, pursuing research and teaching in systems topics together. @zistvan.bsky.social

081