Future of HPC Takes Shape in Monterey
September 19, 2023
By Elizabeth Ball
Leaders in the data science community gathered on the California coast in August, putting their heads together at the annual Monterey Data Conference. Hosted by the Association for High Speed Computing and organized by a committee led by Lawrence Berkeley National Laboratory (Berkeley Lab), the conference is an invitation-only meeting convening researchers from U.S. Department of Energy (DOE) labs and facilities, universities, and industry to exchange ideas about the latest advances and challenges in scientific data analysis and computing.
The theme of the meeting was Connections, an exploration of increasing integration between science, data, and society: cross-facility workflows and integrated research infrastructure (IRI), AI for connected science, data management and processing at scale, and more.
DOE Advanced Scientific Computing Research (ASCR) program manager Ben Brown gave the keynote address on the past, present, and future of IRI, the seamless connection of experimental, data, and compute facilities for science. With the groundwork laid out in the Integrated Research Infrastructure Architecture Blueprint Activity report released by DOE earlier this year, IRI could transform the future of research with real-time analysis of experimental data and opportunities to further incorporate machine learning and AI.
Additionally, Kevin Gomes from the nearby Monterey Bay Aquarium Research Institute reported on integrated data management and its connection to ocean science in his talk “Diving Through a Sea of Data to Understand the Beauty and Importance of Our Oceans.”
Continuing with the theme of connectivity, ESnet, the high-speed network for science based at Lawrence Berkeley National Lab, presented its work developing the next generation of networks to connect facilities worldwide. And researchers from a variety of fields and institutions discussed how they’ve used connected facilities: connected projects such as ESS-DIVE and DIII-D, wildfire prevention and response, and particle accelerators.
This year’s conference was organized by NERSC systems engineer Johannes Blaschke, who said this type of annual gathering is key to fostering learning and teamwork across the scientific landscape. For the first time, this year’s program also included a poster session promoting the work of early-career researchers, just one point of connection over the course of the week. “The hallway conversations, panel discussions, and Q&As were a fantastic way to see what folks from the different labs and domains were thinking,” said Blaschke.
Looking back on the gathering, he said the opportunity to share information and swap ideas is key to the future of HPC for science. “Our mission at NERSC and beyond is to enable scalable and productive team science. This requires that we regularly explore the connections between different scientific domains, projects, facilities, and technologies,” said Blaschke. “Based on what we’ve discussed, sharing and joint analysis of data is what has been driving these connections.”
About Computing Sciences at Berkeley Lab
High performance computing plays a critical role in scientific discovery. Researchers increasingly rely on advances in computer science, mathematics, computational science, data science, and large-scale computing and networking to increase our understanding of ourselves, our planet, and our universe. Berkeley Lab’s Computing Sciences Area researches, develops, and deploys new foundations, tools, and technologies to meet these needs and to advance research across a broad range of scientific disciplines.