INCREASING ADOPTION OF HPC MODELING AND SIMULATION IN THE ADVANCED MANUFACTURING AND ENGINEERING INDUSTRIES

Please Note that a Letter of Intent is due Tuesday, September 05, 2017

PROGRAM AREA OVERVIEW: OFFICE OF ADVANCED SCIENTIFIC COMPUTING RESEARCH

Over the past 30 years, the Department of Energy’s (DOE) supercomputing program has played an increasingly important role in the scientific discovery process by allowing scientists to create more accurate models of complex systems, simulate problems once thought to be impossible, and analyze the increasing amount of data generated by experiments. Computational Science has become the third pillar of science, along with theory and experimentation. Despite the great potential of modeling and simulation to increase understanding of a variety of important engineering and manufacturing challenges, High Performance Computing (HPC) has been underutilized.

Application complexity, in both the development and execution phase requires a substantial in-house expertise to fully realize the benefits of the software tool or service. High capital equipment and labor costs can severely limit a company’s ability to incorporate HPC into their development process. It should also be recognized that changes in HPC hardware including many-core, multi-core processors, GPU based accelerators, and multi-level memory subsystems have made a significant impact on the HPC systems performance and usability. Programming tools and services that can hide this hardware complexity without impacting performance are required.

This topic is specifically focused on bringing HPC solutions and capabilities to the advanced manufacturing and engineering market sectors.

Grant applications are sought in the following subtopics:

a. Turnkey HPC Solutions for Manufacturing and Engineering
HPC modeling and simulation applications are utilized by many industries in their product development cycle, but hurdles remain for wider adoption especially for small and medium sized manufacturing and engineering firms. Some of the hurdles are: overly complex applications, lack of hardware resources, inability to run proof of concept simulations on desktop workstations, solutions that have well developed user interfaces, but are difficult to scale to higher end systems, solutions that are scalable but have poorly developed user interfaces, etc. While many advances have been made in making HPC applications easier to use they are still mostly written with an expert level user in mind.

Grant applications that focus on HPC applications that could be utilized in the advanced manufacturing supply chain including Smart Manufacturing are encouraged as well as applications that address the need to have solutions that are easier to learn, test and integrate into the product development cycle by a more general user (one with computational experience, but not necessarily an expert). Of particular interest are also HPC applications that address engineering challenges related to the design, integration, and fabrication of new devices for Beyond Moore’s Law computing technologies including quantum computing. Issues to be addressed include, but are not limited to: Developing turn-key HPC application solutions, porting HPC software to platforms that have a more reasonable cost vs. current high end systems (this could also include porting to high performance workstations (CPU/GPU) which would provide justification for the procurement of HPC assets, small scale clusters, hybrid platforms or to a “cloud” type environment or service), HPC software or hardware as a service (hosted locally or in the “cloud”), near real time modeling and simulation tools, etc.

Questions – Contact: Ceren Susut, Ceren.Susut-Bennett@science.doe.gov

b. Hardening of R&D Code or Software Tools for Industry Use
The Office of Science Office of Advanced Scientific Computing (ASCR) has invested millions of dollars in the development of HPC software in the areas of modeling and simulation, solvers, and tools. Many of these tools are open source, but are complex “expert” level tools. The expertise required to install, utilize and run these assets poses a significant barrier to many organizations due to the levels of complexity built into them to facilitate scientific discovery and research, but such complexity may not necessarily be required for industrial applications.

Grant applications are specifically sought that will take a component or components of codes developed via the Scientific Discovery through Advanced Computing (SciDAC) program, or other ASCR programs, and “shrink wrap” them into tools that require a lower level of expertise to utilize. This may include Graphical User Interface Designs (GUIs), simplification of user input, decreasing complexity of a code by stripping out components, user support tools/services, or other ways that make the code more widely useable. Applicants may also choose to harden the codes developed by other projects provided that the potential industrial uses support the DOE mission. In addition applicants may choose to strip out code components, harden them and join them with already mature code tools and/or suites of tools to increase the overall toolset and scalability of commercial software.

Questions – Contact: Lucy Nowell, Lucy.Nowell@science.doe.gov

c. Other
In addition to the specific subtopics listed above, the Department invites grant applications in other areas that fall within the scope of the topic description above.

Questions – Contact: Ceren Susut, Ceren.Susut-Bennett@science.doe.gov

Note: In addition to local, cluster, or cloud computing resources, applicants may consider using one of the following DOE Open Science Computing facilities, the National Energy Research Scientific Computing Center, the Argonne Leadership Computing Facility, or the Oak Ridge Leadership Computing Facility.

• Applicants wishing to run at the National Energy Research Scientific Computing Center (http://www.nersc.gov) facility should send email to consult@nersc.gov and inquire about the Education/Startup allocation program.
• Descriptions of the allocation programs available at the Argonne Leadership Computing Facility can be found at http://www.alcf.anl.gov/user-guides/how-get-allocation . Questions concerning allocations this facility can be sent to David Martin dem@alcf.anl.gov. Proprietary work may be done at this facility using a cost recovery model.
• Descriptions of the allocation programs available at the Oak Ridge Leadership Computing Facility are available at http://www.olcf.ornl.gov/support/getting-started/. Questions concerning allocations on this facility can be sent to Jack Wells wellsjc@ornl.gov. Proprietary work may be done at this facility using a cost recovery model.

References: Subtopic a:

1. Kirkley, J., 2011, Making Digital Manufacturing Affordable: A Vendor Perspective, EnterpriseTech. https://www.enterprisetech.com/2011/06/14/making_digital_manufacturing_affordable_a_vendor_p erspective/
2. Executive Office of the President National Science and Technology Council, 20102, A National Strategic Plan for Advanced Manufacturing, p. 51. https://energy.gov/eere/amo/downloads/national-strategic-plan-advanced-manufacturing
3. 2012, Special Report: What is SMART Manufacturing, Time Magazine, p.6. https://www.rockwellautomation.com/resources/downloads/rockwellautomation/pdf/aboutus/company-overview/TIMEMagazineSPMcoverstory.pdf
4. Executive Office of the President, National Science and Technology Council, Committee on Science and Committee on Homeland and National Security of the National Science and Technology Council, 2016, Advancing Quantum Information Science: National Challenges And Opportunities, Interagency Working Group on Quantum Information Science of the Subcommittee on Physical Sciences, p. 23. https://www.whitehouse.gov/sites/whitehouse.gov/files/images/Quantum_Info_Sci_Report_2016_07 _22%20final.pdf

References: Subtopic b:

1. McIntyre, C., 2009, US Manufacturing-Global Leadership Through Modeling and Simulation, High Performance Computing Initiative, Council on Competitiveness, p. 4. http://www.compete.org/storage/images/uploads/File/PDF%20Files/HPC%20Global%20Leadership%2 0030509.pdf
2. U.S. Department of Energy, Scientific Discovery through Advanced Computing (SciDAC). https://science.energy.gov/ascr/research/scidac/
3. U.S. Department of Energy, Scientific Discovery through Advanced Computing (SciDAC). http://www.scidac.gov