Imagine trying to find one specific sentence in a million random books, but the books are scattered across different libraries. Instead of carrying all those books to one place to read them, this system sends a smart assistant to each library to find and summarize only the relevant parts. It makes searching through giant piles of medical images and videos much faster and keeps the data private.
Up to 90% of data is unstructured and locked in bulk storage, making it nearly impossible for data scientists to mine efficiently. Current systems struggle with the massive volume of genomics data and the extreme speed required for real-time surgical video analytics.
A near-data processing platform featuring XtremeDataHub for serverless data connectors, stream operators for low-latency video, and a Data Broker for secure, confidential data sharing.
If you are a genomics research lab dealing with massive text-based genetic data that is hard to move and search — this project developed XtremeDataHub that filters and processes data where it lives. This reduces the need to move bulk data, resulting in higher overall throughput.
If you are a surgical robotics manufacturer dealing with high-speed video streams that require instant analysis — this project developed stream data connectors. These allow for very fast computations over low-latency video and event streams.
If you are a hospital data provider dealing with strict privacy laws that prevent sharing sensitive patient records — this project developed a Data Broker service using Trusted Execution Environments. This allows for secure data orchestration and confidential processing across different locations.
Based on available project data, no specific commercial pricing or cost model is mentioned; it is currently an EU-funded research project.
Yes, the project is designed for 'extreme data' and 'large scale' object storage, specifically targeting high-volume genomics and real-time surgical video streams.
Based on available project data, specific IP and licensing terms are not provided, though the consortium includes 4 industry partners and 1 SME.
The platform uses serverless data connectors to optimize operations like partitioning and filtering directly from Object Storage to analytics services.
The project period runs from 2023-01-01 to 2025-12-31, suggesting the technology is still in development phase.
The consortium is well-balanced for technology transfer, featuring 11 partners across 5 countries. With an industry ratio of 36% (including 4 industry partners and 1 SME), there is significant commercial interest in the outcomes, while the 7 university and research partners provide the deep technical expertise needed for the complex data mining tasks.
Contact Universitat Rovira i Virgili in Spain
Contact us to explore licensing opportunities for the XtremeDataHub connectors.
