The National Science Foundation (NSF) has awarded $4 million to researchers at the Illinois Institute of Technology to address challenges in managing data complexity and scale in scientific research. The funding is part of a larger $5 million grant for the IOWarp project, led by Xian-He Sun, Ron Hochsprung Endowed Chair of Computer Science, and Antonios Kougkas, assistant research professor of computer science.
IOWarp seeks to optimize data transfer through innovative techniques and provide a unified platform capable of handling diverse data sources and formats. This initiative aims to simplify data management for scientists across various fields.
“I’m particularly energized by how IOWarp isn’t just another generic data platform,” Kougkas states. “It’s been carefully designed through direct collaboration with scientists across diverse fields such as materials science, cosmology, and biomedicine.”
The project builds on previous work by Sun and Kougkas on Hermes, a multi-tiered distributed input/output system. IOWarp incorporates features from Hermes while adding new functionalities tailored for modern scientific workflows that include artificial intelligence applications.
“This is the third multi-million-dollar NSF grant that our group has received over the last six years,” Sun comments. “From the Hermes data management and transfer systems for high performance computing, to the ChronoLog data systems for cloud computing, to the current IOWarp system for AI applications... This award is a great recognition for us and for Illinois Tech.”
IOWarp promises advancements in three key areas: managing diverse data types across workflow stages, reducing transferred data volume via tiered content organization, and supporting varied data sources.
“When I think about how IOWarp could accelerate breakthrough discoveries... it’s hard not to be enthusiastic about the impact this could have on scientific progress as a whole,” says Kougkas.
A notable feature of IOWarp is its natural language interface powered by WarpGPT. This AI-driven technology allows scientists to explore complex datasets using natural language queries instead of code, thus democratizing access to sophisticated analyses.
“IOWarp’s natural language interface is not just a tool—it’s a vision for the future of scientific data management,” adds Kougkas.
Another unique aspect is IOWarp's "content" representation which standardizes varied data inputs into formats any application can process. This innovation simplifies integration with existing tools while easing transitions from established platforms—a significant challenge acknowledged by Kougkas: “The biggest hurdle...is convincing the scientific community to adopt this new platform.”
To overcome this challenge, building an active open-source community around IOWarp will be crucial. Such engagement would drive ongoing development ensuring relevance amidst rapid technological changes in scientific computing.
“As someone passionate about advancing scientific discovery...the open-source nature of the project adds another layer of excitement,” concludes Kougkas.
Sun and Kougkas collaborate with researchers at the University of Utah and The HDF Group on developing IOWarp software. Interested parties can follow developments or join their GitHub community efforts online.
Research reported here receives support under NSF Award Number 2411318; content responsibility lies solely with authors without representing official NSF views.
___