The High Performance Data Facility (HPDF) is a first-of-its-kind distributed, hybrid facility. The HPDF team is building a state-of-the-art user facility and resource for scientific research. HPDF is a partnership between Thomas Jefferson National Accelerator Facility (Jefferson Lab) and Lawrence Berkeley National Laboratory (Berkeley Lab). Core HPDF infrastructure will be sited at each lab’s campus providing resilient data infrastructure to the SC community. 

The mission of HPDF is to enable and accelerate scientific discovery by delivering expert data management infrastructure, capabilities, and tools. Our guiding vision is enabling scientists and AI systems to make new scientific discoveries by seamlessly working with distributed and federated datasets. As part of the ASCR Facilities Ecosystem, HPDF will provide essential data management capabilities, including hardware, software, and experts to manage the large-scale scientific data produced at DOE user facilities and beyond. HPDF will be a critical resource for enabling the frontier of AI-enabled scientific discovery. HPDF is an Infrastructure Partner (IP) in the American Science Cloud. The HPDF Project is sponsored and supported by the Department of Energy’s Office of Science’s (DOE SC) Advanced Scientific Computing Research (ASCR) program.

How HPDF Will Impact Scientific Communities

HPDF is building a cutting-edge high-performance data platform that provides advanced storage, processing, and analytical services, empowering data-intensive research and AI-driven discoveries.

Partner with us

Interested in partnering with the LBNL HPDF Project team?

We are engaged in ongoing conceptual design and early data and AI services development for the HPDF Project and as core contributors to American Science Cloud. Public reports outlining insights and design concepts can be found here. We are happy to engage and learn more about your community’s scientific data challenges to identify areas for collaboration. These opportunities are essential to shaping our conceptual design plus designing and delivering usable services as rapidly as our resources allow. Partnership opportunities with the LBNL HPDF Project team include:

Scientific Users — Access and test the range of hardware, software, and services supported by HPDF.

Co-Design Partnerships — Longitudinal engagements with science teams building, evaluating and supporting community data resources to achieve mutually beneficial advancements.

The HPDF Berkeley Lab Team

The HPDF team at Berkeley Lab is located in Shyh Wang Hall. Berkeley Lab’s HPDF team is committed to supporting scientific research patterns and data stewardship within the nation’s research communities.

The Berkeley HPDF team includes:

Leadership

Support

  • Jason Salinas, Administrative Support
  • Ingrid Ockert, Communications

Technical Team

HPDF Project Status & Plan

This overview of the HPDF project explores the origins of the project and its mission. It also provides information about its intended physical design, infrastructure design, and scope.

Initial User Study Insights

The HPDF User Experience (UX) team conducted eight semi-structured interviews with twelve individuals leading data and computing infrastructure work at the DOE Office of Science (SC) user facilities and projects. This report presents key takeaways synthesizing community perspectives and needs, along with recommendations for the project. 

Community Requirements Meta-Analysis

This meta-analysis examines the needs of the breadth of the SC community, captured in publicly available community reports or mission documents. The meta-analysis identifies and provides initial characterization of fifteen core requirements for the HPDF Project team to consider during the conceptual design phase. 
Last edited: August 28, 2026