Keynote
Raghu Ramakrishnan
CTO for Azure Data and Technical Fellow at Microsoft
Evolving Data Lakes for an Agentic Future
About the Workshop
The goal of FORMATS is to bring researchers and practitioners together to shape the future of data formats. It offers a forum for sharing ideas, exploring new directions, and accelerating progress toward standardization and adoption. The workshop rethinks traditional database research to meet the needs of modern workloads and infrastructure, with topics spanning table and file formats, data encodings, operational efficiencies on modern hardware, and their role in analytics, transactions, vector processing, and AI applications.
We believe that FORMATS can foster greater collaboration between academia and industry, and help chart new research directions, adoption paths, and standardization for table, file, and data formats.
Program Highlights
Room: Scarlet 2
Invited Talk
Sumedh Sakdeo
Principal Staff Software Engineer at LinkedIn
Full-Stack Lakehouse Architecture at Exabyte Scale: Lessons and Research Opportunities from LinkedIn's OpenHouse-Managed Iceberg
Invited Talk
Pedro Pedreira
Software Engineer at Meta
Scaling AI Training: Storage Layout Challenges
Invited Talk
Will Manning
Co-founder/CEO at Spiral
Turtles All the Way Down: Composability & File Formats
Panel Discussion
Does One Size Still Fit All? Data Formats in a Multi Engine, Agentic Stack
Moderator: Ashvin Agrawal
Workshop Organizers
Program Committee
- Azim Afroozeh — CWI
- Chris Douglas — UC Berkeley
- Chunwei Liu — Purdue University
- Jana Giceva — TU Munich
- Jesús Camacho-Rodríguez — Microsoft
- Martin Prammer — CMU
- Mosha Pasumansky — Firebolt
- Virajith Jalaparti — Databricks
- Weston Pace — LanceDB
- Will Manning — Spiral
- Yuanyuan Tian — Microsoft Gray Systems Lab