Award Abstract # 2139536
Characteristic Science Applications for the Leadership Class Computing Facility

NSF Org: OAC
Office of Advanced Cyberinfrastructure (OAC)
Recipient: UNIVERSITY OF TEXAS AT AUSTIN
Initial Amendment Date: August 26, 2021
Latest Amendment Date: July 17, 2024
Award Number: 2139536
Award Instrument: Cooperative Agreement
Program Manager: Edward Walker
edwalker@nsf.gov
 (703)292-4863
OAC
 Office of Advanced Cyberinfrastructure (OAC)
CSE
 Directorate for Computer and Information Science and Engineering
Start Date: September 1, 2021
End Date: January 31, 2025 (Estimated)
Total Intended Award Amount: $6,999,361.00
Total Awarded Amount to Date: $6,999,361.00
Funds Obligated to Date: FY 2021 = $6,999,361.00
History of Investigator:
  • Daniel Stanzione (Principal Investigator)
    dan@tacc.utexas.edu
  • John West (Co-Principal Investigator)
  • Omar Ghattas (Co-Principal Investigator)
  • John Cazes (Co-Principal Investigator)
Recipient Sponsored Research Office: University of Texas at Austin
110 INNER CAMPUS DR
AUSTIN
TX  US  78712-1139
(512)471-6424
Sponsor Congressional District: 25
Primary Place of Performance: University of Texas at Austin
3925 W Braker Lane, Suite 3340
Austin
TX  US  78759-5316
Primary Place of Performance
Congressional District:
37
Unique Entity Identifier (UEI): V6AFQPN18437
Parent UEI:
NSF Program(s): Leadership-Class Computing
Primary Program Source: 01002122DB NSF RESEARCH & RELATED ACTIVIT
Program Reference Code(s):
Program Element Code(s): 778100
Award Agency Code: 4900
Fund Agency Code: 4900
Assistance Listing Number(s): 47.070

ABSTRACT

The goal of this project is to select, study, and transform a set of Characteristic Science Applications (CSAs) to enable next-generation science on the NSF Leadership-Class Computing Facility (LCCF). The CSAs were chosen to represent a broad range of science domains and computational approaches, and each CSA will comprise one or more computer codes and a challenge problem to be solved on the LCCF. The set of CSAs will enable the project to verify the design of the LCCF and validate that the facility is applicable across the broad range of science disciplines supported by the NSF when it is constructed. Furthermore, the transformations made to the codes and workflows in the CSAs will provide best practice models and implementation exemplars that will inform training materials, tutorials, and other educational content. This will ensure that the nation's science and engineering research community will be effective on the LCCF from day one of operations.

The project plans to evaluate a broad selection of science applications and challenge problems, laying the foundations for a transformation of the computer codes in a way that will support the LCCF broad goal of enabling a ten-fold or more time to solution performance improvement over NSF's current leadership computing resource, Frontera. This capability improvement will come from a combination of increased scale of the LCCF primary computing system(s), increased capabilities of the components (on-node I/O bandwidth, inter-node I/O bandwidth, disk and memory bandwidth, access speed, cores, etc.) that make up the computing instrument, and improvements in the software codes that make up the set of science applications used to demonstration ten-fold performance improvement.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

PROJECT OUTCOMES REPORT

Disclaimer

This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.

 

The Characteristic Science Applications (CSA) project is closely coordinated with the design of the NSF Leadership Class Computing Facility (LCCF), the NSF's next large scale investment in computational infrastructure for science, engineering, and AI. The CSA project's goals were to select a set of "challenge" problems for the new facility to:
Inform the design of the LCCF supercomputer
Build a benchmark suite to clearly demonstrate the performance of the new system (with a goal of having 10x the capability of the previous generation).
Prepare the community to run on this new supercomputer, at scale with high impact science on day 1.
This project was effective in all 3 of its goals. More than 150 science teams applied to be part of the initial cohort of projects. Ultimately, 20 teams were selected to be funded partners on the project, with problems spanning biology, astronomy and astrophysics, materials science, extreme weather events, seismological monitoring, machine learning, and others. These 20 projects and their application workflows helped decide the architecture for the next LCCF system, Horizon, then work turned to porting and tuning the software to be ready for the new system. At the end of the project, these 20 were downselected to 11 applications to move forward in the Construction phase of the LCCF award, though nearly all the 20 are ready for use on Horizon immediately when it is deployed on 2026. Performance projections on these codes also show that Horizon will exceed its performance targets, with an expected average closer to 15x rather than 10x the performance of the existing NSF systems. As expected, the performance improvement is equally due to the new hardware the system will provide, and algorithmic and software improvements enabled via CSA.    

 


Last Modified: 06/02/2025
Modified by: Daniel Stanzione

Please report errors in award information by writing to: awardsearch@nsf.gov.

Print this page

Back to Top of page