The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Authors: Jeremy Crampton (U.S. Dept. of Energy), Sadaf Alam (University of Bristol), Verónica Melesse Vergara (Oak Ridge National Laboratory (ORNL)), Debbie Bard (Lawrence Berkeley National Laboratory (LBNL)), Aleksi Kallio (CSC - IT Center for Science), Benjamin Brown (US Department of Energy), Feiyi Wang (Oak Ridge National Laboratory (ORNL)), Rafael Ferreira da Silva (Oak Ridge National Laboratory (ORNL)), Carol Hawk (DOE Office of Advanced Scientific Computing Research), Arjun Shankar (Oak Ridge National Laboratory (ORNL)), Jason Haga (National Institute of Advanced Industrial Science & Technology)
Abstract: We explore the multifaceted aspects of managing risk for research projects involving sensitive data and AI models which depend deeply on supercomputing infrastructures. This risk spectrum spans traditional technical cyber controls as well as policy and sociological (including human factors) risks. In the context of multi-facility, multi-institutional workflows such as IRI and the American Science Cloud (AmSC), our goal is to advance progress in developing secure and trustworthy infrastructures for AI and integrated science.
Three intertwined challenges emerge as we advance this vision for IRI and AmSC: technological, policy, and sociological. With the rise of AI and the increasing use of sensitive data for training models, our goal is to leverage this BoF to build a community of practice that will advance a secure and trusted research environment (TRE) that addresses challenges in all three domains. How do we best achieve a TRE that is transparent, reproducible, ethical, secure, worthwhile, and collaborative, with clear data provenance and assurance? How might trust be rightfully earned and retained through modern workflows through managed risk and secure governance?
The outcomes from this BoF are: (1) explore TRE challenges in the age of AI and science integration; (2) identify alignment and divergence in TRE practices; (3) learn from complementary efforts across institutions around the globe; and (4) build a community of practice committed to trustworthy integrated science in the age of AI. We invite an audience with interests in these topics to participate in advancing these outcomes.
Long Description: This BoF explores the multifaceted aspects of managing risk for multi-facility research projects in today’s rapidly changing environment of AI and integrated science workflows that may require sensitive data. AI and integrated science make new demands on making supercomputing infrastructures secure and trustworthy. As the 2024 ASCAC Report emphasizes, linking distributed resources is paramount to modern collaborative science which demands an ecosystem that offers more than the sum of its parts. The report concludes that, to maintain scientific leadership, “business as usual” cannot continue. The Department of Energy’s vision for Integrated Research Infrastructure (IRI) empowers researchers to meld DOE’s world-class research tools, infrastructure and user facilities seamlessly and securely in novel ways to radically accelerate discovery and innovation.
Three intertwined challenges emerge as we advance this vision for IRI: technological, policy, and sociological. With the rise of AI and the increasing use of sensitive data for training models, our goal is to leverage this BoF to build a community of practice that will advance a secure and trusted research environment (TRE) that addresses challenges in all three domains.
Trust requires humans to accept a system as transparent, reproducible, ethical, and collaborative (eg., as described in the Turing Way, a project to enable open science, open collaboration, and community-driven research), with clear data provenance and assurance., As such it is a sociological, policy, and technical achievement. Trust involves the acceptance of managed risk. Scientists choosing to participate in the TRE will weigh the benefit of accelerating scientific discovery through multi-facility integrated research with the residual risk that remains after thoughtful risk enumeration and mitigation, which this BoF seeks to further advance. Additionally, we need to consider trust throughout the workflow; what is it that is trusted–the technology; the data; the management; or the results?
This BoF will explore critical challenges that DOE and its national laboratories, partner federal agencies, and international supercomputing centers are addressing as we work to create an open innovation system; accelerate discovery and innovation; democratize access; and advance open and findable, accessible, interoperable and reusable (FAIR) science, within a trustworthy IRI ecosystem.
The outcomes from this BoF are: (1) explore TRE challenges in the age of AI; (2) identify alignment and divergence in TRE practices (3) learn from complementary efforts across institutions around the globe; (4) build a community of practice around trustworthy integrated science in the age of AI.
We have confirmed participation from Sadaf Alam, Director of Advanced Computing Strategy, University of Bristol, UK; Pekka Manninen, Director of Science and Technology, CSC – IT Center for Science, Finland; Deborah Bard, NERSC Science Engagement and Workflows Department Head, LBNL; Veronica Vargara, ORNL; Arjun Shankar, Division Director, National Center for Computational Science, ORNL; and Ben Brown, Director of Facilities, ASCR, DOE. Additionally, we have invited Jason Haga, Chief Senior Research Scientist at National Institute of Advanced Industrial Science and Technology (AIST), Japan.
Website: https://docs.google.com/document/d/1F7DRtRcLe-0QSYUJmAOizki8F2yrhb4Ld-EDXl-5Vt0/edit?usp=sharing