By Scott Sundt, High Performance Computing Modernization Program, Alexandra Landsberg, High Performance Computing Modernization Program, Megan Holland, ERDC and Owen Eslinger, ERDC
The rollout of the National Defense Strategy in early 2018 had an immediate impact on the way DOD approaches its mandate for defending the United States. It has sent ripples throughout the services and caused much focused reflection on the strategic, operational and tactical priorities that must be addressed in a future world where potential near-peer adversaries may have reached parity with our own force structure. Within DOD research, development, test and evaluation communities, the call has gone out to speed up innovation, reduce acquisition timelines, and produce game-changing weapon systems at reduced cost and risk. The directive from the National Defense Strategy is clear: "Deliver performance at the speed of relevance." The American warfighter needs 21st-century weapons systems now, not 10 or 15 years from now.
Fortunately, DOD finds itself at a unique moment thanks to the steady progress of digital engineering and high-performance computing over the past 20 years. In response to increasing access to supercomputing, massive data collection, high-fidelity physics-based software development, and high-speed networks, DOD released its Digital Engineering Strategy in June 2018. This document has set a course for DOD and the services to take advantage of these enabling assets and apply them to the emerging technology challenges.
The combination of the National Defense Strategy and the Digital Engineering Strategy has provided the Army with the impetus for fundamentally restructuring the way it equips its forces for the future. The Army's top six modernization technologies (long-range precision fires, Next Generation Combat Vehicle, Future Vertical Lift, the network, anti-missile defense and Soldier lethality), along with its eight cross-functional teams and a new four-star U.S. Army Futures Command, have provided a solid foundation and direction for the Army research, development, test and evaluation community and program executive offices.
HIGH PERFORMANCE COMPUTING PROGRAM
The DOD High Performance Computing Modernization Program provides high-performance computing capabilities and expertise, enabling National Defense Strategy priorities across DOD. It provides DOD with a comprehensive computational modeling and simulation ecosystem that integrates supercomputing capabilities, high-speed networks and computational science expertise that enable scientists and engineers to conduct a wide range of focused research, development and test activities.
The program, with a $280 million annual budget, was chartered by Congress to revolutionize warfighter support through increased application of high-performance computing to critical research, development, test and evaluation and acquisition engineering initiatives. This is a triservice effort managed for the assistant secretary of the Army for acquisition, logistics and technology (ASA(ALT)) by the deputy assistant secretary of the Army for research and technology and executed by the U.S. Army Engineer Research and Development Center (ERDC) in Vicksburg, Mississippi.
By amplifying the creativity, productivity and impact of the research, development, test and evaluation and acquisition engineering communities, high-performance computing provides unprecedented insight into the physical world that would be too costly, dangerous or time-consuming to obtain through observation and experimentation alone. The program includes DOD supercomputing resource centers; software applications; and secure networking. The High Performance Computing Modernization Program also leverages specialized expertise from DOD, other federal departments and agencies, industry and academia to mature leading-edge software application codes. This expertise complements that of DOD scientists and engineers, helping customers achieve critical mission objectives.
As the demands and availability for computational resources have grown over the years, the user base and the size and complexity of jobs have grown to take advantage of the opportunities afforded by increased resources. In fiscal year 2019, the High Performance Computing Modernization Program internal database indicated there were over 3,000 active users. Large-scale high-performance computing simulations have transformed from a niche activity to a mainstream activity.
The software applications of the High Performance Computing Modernization Program provide a suite of software development and support services aimed at optimizing software capabilities to design, develop, test and deploy superior DOD capabilities. These efforts and services include the Computational Research and Engineering Acquisition Tools and Environments (CREATE) software development activity, which furnishes critical modeling support in the world of digital engineering prototyping.
CREATE is the program's premier vehicle for addressing DOD's current and future design and analysis efforts for its major acquisition programs. CREATE provides innovative applications of its software tools for developing and optimizing aircraft, ship, ground vehicle and radar antenna designs, and allows the acceleration of the acquisition timeline. It has expanded the acceptance, use and adoption of its various physics-based software tools to over 160 defense organizations. Its products are now becoming an integral part of major defense acquisition programs for design space exploration, design analysis, and performance prediction and testing across the weapon system life cycle.
Present acquisition programs largely follow an empirical "design-build-test" iterative methodology. This results in late discovery of design flaws, issues of immature technology and system integration problems. Rework and redesign efforts contribute substantially to cost overruns and schedule delays. By employing a "model-test-build" paradigm, optimized engineering designs can be developed early in the acquisition process using CREATE tools. Costs can be substantially reduced; schedules shortened; and design and program flexibility and agility increased. Above all, the reduction of design flaws, the quick and flexible development of sound engineering concepts and designs, and beginning the systems integration engineering much earlier in the acquisition process all improve the performance of acquisition programs.
IMPACT ON ARMY PROGRAMS
High-performance computing supports the entire life cycle of a weapon system. Using the Army's helicopters as an example, high-performance computing is critical to the updating of legacy platforms such as the CH-47 Chinook and the H-60 Black Hawk, as well as the Army's Future Vertical Lift effort. CREATE-AV's (Aviation Vehicles) Helios software is a high-fidelity, multi-physics analysis tool for rotary-wing aircraft. Helios can calculate the performance of a full-sized rotorcraft, including the fuselage and rotors. It can also handle arbitrary rotor configurations, and analyze and predict prescribed maneuvers with tight coupling of rotor aero-structural dynamics. A highly accurate treatment of the complex air flow generated from rotor blade tips-vortex shedding-gives Helios the unique capability to assess the interaction of these vortices with the fuselage and nearby rotor blades. The large-scale calculations with Helios are run on the program's supercomputers. Helios offers the ability to predict phenomena that, a decade ago, could only be observed in flight test.
The CH-47 Block II Advanced Chinook Rotor Blade is designed for improved lifting capability in hover without compromising forward speed. Initial flight tests showed high-control system loads for the rear rotor in high-speed forward flight. Army engineers, Boeing Co. and the Project Manager for Cargo Helicopters under the Program Executive Office for Aviation formed an engineering team to address this problem. Helios was used to capture the complex, unsteady aerodynamics phenomena and explore design space to restore high-speed performance while retaining the modified rotor blade's benefits for hover. Army engineers have been able to identify potential performance issues and evaluate mitigation designs. The modified rotor blade design was successfully tested in late 2018, resulting in a significant enhancement to the combat capability of the 400-plus Chinooks in the Army inventory. In testimony to the Senate Armed Services Airland Subcommittee, Lt. Gen. Paul A. Ostrowski, principal military deputy to the ASA(ALT), highlighted the importance of the High Performance Computing Modernization Program to Army acquisition programs.
"It is absolutely critical," said Ostrowski. "With respect to the Block II Chinook (helicopter), we have avoided about $50 million of cost in terms of flight based on being able to supercompute the effects of the new rotor blades."
The Joint Multi-Role Technology Demonstrator (JMR-TD) program is a precursor to the Army's Future Vertical Lift effort, intended to demonstrate transformational vertical lift capabilities to enable programmatic decisions. Requirements for the JMR-TD aircraft were established in 2012. In 2013, technology investment agreements were awarded to four companies-AVX Aircraft Co., Bell Textron Inc., Karem Aircraft Inc. and Sikorsky Aircraft Corp. teamed with Boeing. In an effort to better understand the aircraft and technologies being developed under JMR-TD, Helios was applied to all four of the configurations awarded under the technology investment agreements in 2013. This enabled Army aviation engineers to conduct an independent analysis of contractor proposals, resulting in more informed and timely acquisition decisions. Results from the analysis were used during the initial design and risk review to guide selection of the two demonstrator aircraft from Bell and the Sikorsky-Boeing team. Helios continues to be used to carry out further analysis on the two test-bed aircraft selected for development. Because the actual vehicles have proprietary information, these results cannot be disseminated openly.
The High Performance Computing Modernization Program also works closely with the Engineered Resilient Systems program, an effort initiated in 2012 to accelerate acquisition through the use of high-performance computing. The Engineered Resilient Systems approach combines high-fidelity, physics-based modeling, advanced data analytics, machine learning and process automation with high-performance computing to enable better-informed decisions before major acquisition milestones. The partnership with Engineered Resilient Systems allows Army high-performance computing to be used in new ways; the program recently developed an artificial intelligence and machine learning ecosystem for large-scale data management. As a result, the Army was able to consolidate terabytes of H-60 helicopter data, giving analysts the ability to explore the full maintenance dataset and enabling true predictive maintenance for the first time. When fully implemented, 100 percent of the H-60 fleet will be eligible for the oil cooler life extension program versus the 20 percent previously eligible. This has the potential to double the maintenance interval for the oil cooler, a critical component that costs several hundred thousand dollars to replace. This practice is already being extended to fleets of ground vehicles and other Army platforms.
The High Performance Computing Modernization Program is a national asset delivering high-performance computing capabilities and expertise to mission-critical challenges. Together with the Engineered Resilient Systems program, it is striving to improve acquisition efforts across all phases of the weapon system life cycle and to enable more informed and timely acquisition decisions. These programs have already demonstrated the ability to accelerate the acquisition timeline, while also reducing risk and cost to DOD, and will continue to engage in new partnerships to address the department's highest priorities.
SCOTT SUNDT is a retired Navy captain with over 30 years of active-duty service, including command at sea. He is the lead for High Performance Computing Modernization Program Acquisition and Digital Engineering. He holds an M.S. in electrical engineering from the Naval Postgraduate School, an M.S. in national strategic studies from the National War College and an M.S. in national resource studies from the Industrial College of the Armed Forces of National Defense University, and has a B.S. in physical science from the U.S. Naval Academy.
ALEXANDRA LANDSBERG is the deputy director of the High Performance Computing Modernization Program. She holds an M.S. and a B.S. in aerospace engineering from the Massachusetts Institute of Technology. She has over 25 years of experience with the federal government in high-performance computing.
MEGAN HOLLAND is a knowledge management specialist at the ERDC Information Technology Laboratory in Vicksburg, Mississippi. She has an MBA with an emphasis in marketing from Mississippi State University and a B.A. in English with an emphasis in writing from Mississippi College.
OWEN ESLINGER is the Engineered Resilient Systems program manager and a computer scientist at the ERDC Information Technology Laboratory. He holds a Ph.D. and an M.S. in computational and applied mathematics from the University of Texas at Austin, and a B.S. in mathematics from North Carolina State University.
This article is published in the Winter 2020 issue of Army AL&T magazine.