The platform offers a comprehensive and interactive database that allows users to delve into the software layers of leading AI models
As part of an ongoing eight-year collaboration, the Barcelona Supercomputing Center - Centro Nacional de Supercomputación (BSC-CNS) and Micron Technology have unveiled FAiNDER (Find comprehensive AI data and insights) an open-source web platform designed to transform how researchers and developers navigate the rapidly evolving AI landscape. FAiNDER provides centralized, up-to-date information on the key system requirements of all major AI models to facilitate exploration and optimize hardware choices.
The FAiNDER.eu platform offers a comprehensive and interactive database that allows users to delve into the software layers of leading AI models to help them understand the hardware resources needed for their efficient execution.
With FAiNDER, users choose the depth at which they want to explore the AI world. For students and junior researchers, the tool offers a starting point for exploring any AI model. It provides a high-level view of the models’ software architecture and their main features, and it includes a list of reliable sources for future exploration. For advanced users, the tool provides detailed information on all phases of the software pipeline, their main arithmetic operations, data types, memory footprint and the target hardware platforms, enabling quick comparisons and deep technical insights.
Petar Radojkovic, the BSC team leader driving the development of this tool, says, “FAiNDER started as an internal repository for relevant AI studies. It was created to handle the overwhelming volume of publications in this rapidly evolving domain. We gradually enriched the tool until it became the must-have tool for any AI-related exploration and discussion between the BSC and our industrial partners. We now want to open the FAiNDER.eu platform to the whole community.”
“Micron’s high-performance memory and storage solutions are critical to enabling AI and HPC workloads, and the FAiNDER tool provides users with an extensive database of AI models along with the appropriate hardware required to optimize performance,” said Emanuele Confalonieri, technical project leader and distinguished member of technical staff at Micron. “FAiNDER will enable computing architects and the scientific research community to navigate multiple studies and identify the information they need for their specific design or application. We value the close collaboration with BSC to deliver this AI capability and look forward to our continued joint efforts.”
Key Features and Benefits
- Comprehensive database: FAiNDER currently synthesizes data from 136 top-tier research papers from academia and industry. Sources are continuously updated with credible data so researchers, developers and enthusiasts can make informed decisions about AI technologies.
- Customizable exploration: Users can select the depth of exploration, from basic overviews of AI models to detailed breakdowns.
- Timing- saving insights: FAiNDER streamlines the process of discovering and comparing AI models, helping researchers, developers and enthusiasts shave hours off their research and development
FAiNDER has already been used in key research initiatives undertaken by the BSC, and it is poised to become an essential tool for AI research in the future. Additional joint efforts are now underway to address the challenges associated with memory systems bottlenecks in important application domains and to explore advanced memory systems for high-performance computing and AI.
About the BSC-Micron collaboration
The FAiNDER platform is one of the activities stemming from the eight-year collaboration between BSC and Micron. BSC and Micron have been collaborating since 2017 within the framework of a sponsored research agreement. Several notable dissemination activities have been jointly pursued targeting exploration of future memory systems, including: a co-authored publication presented at the 57th IEEE/ACM International Symposium on Microarchitecture (MICRO), the European Memory Systems Forum hosted as part of the EU-funded DEEP SEA project, a publication in Transactions on Computers, the Micron-BSC HiPEAC technology transfer awards in 2020 and 2022 and this FAINDER platform, which was also presented at the 2024 ACM European Summer School on HPC Computer Architectures for AI and Dedicated Applications.