Systems-Group at SIGMOD'26 with Carsten Binnig as Program Chair

2026/06/04

The Systems Group attended the ACM SIGMOD/PODS 2026 Conference in Bengaluru, India, one of the leading international conferences in data management and database systems. SIGMOD brings together researchers and practitioners from academia and industry to discuss the latest advances in database systems, data analytics, AI for data management, and large-scale data processing.

A particular highlight for our group was that Prof. Carsten Binnig served as Program Co-Chair of SIGMOD 2026, helping to shape the scientific program and oversee the review process of the conference.

The Systems Group contributed to the conference and its affiliated workshops with a broad range of presentations spanning data management for AI, AI for data management, secure data processing, and modern hardware-aware database systems. Our presentations included:

  • What No One Tells You About Page Pruning in Parquet: The Real Cost of Page Index Parsing (Formats Workshop)
  • EcoLLM: Energy-Aware Benchmarking of LLMs for Data Processing Workloads (aiDM Workshop)
  • Deal with it! Towards Self-organizing Data Schemas from Semi-Structured Inserts (aiDM Workshop)
  • Secure Multi-Party Analytics in the Enterprise Revisited: Opportunities and Challenges (SeQureDB Workshop)
  • How to Make Secure Storage Fast for DBMSs in Intel SGXv2 (DaMoN Workshop)
  • Do GPUs Really Need New Tabular File Formats? (DaMoN Workshop, Best Short Paper Award)
  • ScaleEvict: Altruistic Eviction for RDMA-Enabled Distributed Storage Engines (DaMoN Workshop, Best Full Paper Award)
  • Boosting DBMS Test Coverage via LLM-Driven SQL Generation (DBTest Workshop)
  • Towards More Realistic Natural Language Queries in Text-to-SQL Benchmarks (HILDA Workshop)
  • Keynote: *Do We Still Need Databases in an AI World? (DEEM Workshop)

We are especially pleased that two Systems Group papers received Best Paper Awards at DaMoN 2026: Do GPUs Really Need New Tabular File Formats? and ScaleEvict: Altruistic Eviction for RDMA-Enabled Distributed Storage Engines.

We thank the organizers, authors, reviewers, and attendees for an inspiring week of discussions and look forward to continuing these conversations at future database conferences.