A growing number of articles describe the challenges of adopting DevOps for DoD software development. A recent paper in this journal discussed DevOps adoption challenges within the acquisition programs of the Navy. For embedded software, these challenges have been overwhelming to date. However, for standalone applications, the story is different, even though the lingering impact of DoD Instruction 5000 and other DoD cultural practices remains problematic. The HPCMP CREATE™ program, or simply CREATE in this paper, is part of the portfolio of the DoD High Performance Computing Modernization Program (HPCMP). It has successfully adapted DevOps concepts and agile software development practices to develop a family of software applications that enable system-scale virtual prototyping and testing analysis for major DoD acquisition programs. This study describes the main enabling practices that made this possible.

CREATE (Computational Research and Engineering for Acquisition Tools and Environments) is a risk-based, practice-centered software development program in support of Depart of Defense (DOD) digital engineering. With the latest releases of Kestrel and Helios, the CREATE program illustrates its latest application of the “software factory” concept and agile DevOps cycle advanced by the Defense Science Board (DSB) [1] and the DOD Office of the Chief Information Officer (CIO) [2]. The companion paper [3] to this article, addresses the “Dev” portion of the DevOps model used for product development. The focus in Part II is on the “Ops” component, to include updates of quality assurance (QA) practices. The overall goals on the operations side of the DevOps cycle are to provide the best end-to-end user experience across all platforms and accelerate engineering workflows supporting DOD acquisition programs through the use of CREATE’s physics-based computational tools.

The HPCMP CREATETM program (“Computational Research and Engineering Tools and Environments Program,” or “CREATE”) is a 13-year old federation of DoD Service (Army, Navy, and Air Force) organizations and federal contractors that are software enablers of physics-based virtual prototypes, digital surrogates and twins, and the digital thread for major DoD military systems such as air vehicles, ships, ground vehicles and radio frequency antennas. The CREATE product development component of the management model can be described as a risk-aware implementation of the Defense Science Board’s DevOps-based “software factory” concept, although CREATE pre-dates the appearance of the terms “DevOps” (2009) and “software factory” (2018). This paper will describe the CREATE implementation of DevOps emphasizing a risk-aware approach to product development. A separate companion paper will address the growing importance of the “Ops” component of the model.

The USA faces a multitude of threats to its national security and international interests in an era of exponential technology growth and unprecedented access by anyone with a smartphone. Traditionally, the acquisition of US defense systems has relied on sequential methods of conceptual design and development. While successful in the past, these methods are time consuming and in danger of creating vulnerability gaps that could limit or constrain US response options. The challenge is clear. Either the US Department of Defense (DoD) evolves the way it plans, develops, buys, and manufactures new weapons systems, or it cedes the high ground to a rapidly changing global environment. Adapting and expanding advances in high performance computing (HPC), developing and employing complex physics-based software tools for high fidelity modeling and simulation, and implementing a vision that combines these elements with other processes are critical enablers the DoD is pursuing. This paper describes the synergy of three major DoD efforts designed to address needs in the areas of acquisition program development and execution: the DoD High Performance Computing Modernization Program (HPCMP) (the DoD HPCMP began as an Office of the Secretary of Defense program in 1992; in October 2011, leadership transferred to the Assistant Secretary of the Army for Acquisition, Logistics and Technology; the US Army Corps of Engineers Engineer Research and Development Center manages the program, https://www.hpc.mil/index.php ) Computational Research and Engineering Acquisition Tools and Environments program; the Engineered Resilient Systems program; and the DoD Digital Engineering vision.

Recent events in the US defense community hold the promise of a major boost in the use of agile software development methods and a concomitant improvement in the effectiveness of defense software procurement and development.

The goal of the CREATE program is to develop and deploy physics-based computational engineering tools that can be used to develop virtual prototypes of ships, air vehicles, ground vehicles, and radio frequency antennas to accurately predict their performance in support of the US Department of Defense acquisition process, DoD 5000. The purpose of this article is to describe the approach taken to address the verification and validation of the CREATE software products. The approach is based on the adoption of a set of practices aligned with the recommendations of the National Academy of Sciences to promote a test-driven development culture.

Silicon Valley is the home of many of the most innovative high-technology industries that have ever existed. Their level of innovation gives them a competitive economic advantage that sustains a significant portion of the US economy. While there have been dozens of attempts in the US and abroad to replicate this success, few have been very successful. A relatively new approach to product development is emerging, computational engineering. It is based on the use of computing, especially high-performance computing, to design, construct, and analyze virtual prototypes of new products.

With an annual budget of nearly US600 billion, the US Department of Defense (DoD) is tasked with protecting the US and its allies and interests abroad against potential adversaries. It must accomplish these tasks in a globalized and highly interconnected world where the pace of technology is moving faster than the current time-consuming DoD acquisition systems can keep up with. Moving forward, the DoD must change these time-consuming paradigms and accelerate the pace of innovation. One way to do this is to shift to a virtual prototyping environment where the benefits of high-performance supercomputing and complex physics-based engineering software can expand the decision space and enable the DoD to field better systems faster, cheaper, and at less risk than previous methods. This is the goal of the Computational Research and Engineering Acquisition Tools and Environments (CREATE) program. This issue of CiSE provides the third installment of articles relating to the history, formation, and ongoing work of the High Performance Computing Modernization Program (HPCMP) CREATE program.

Today, rapid product innovation is essential to remain competitive. To help spur innovation in the acquisition of major defense systems and reduce their cost, time, and risks, the US Department of Defense launched the Computational Research and Engineering Acquisition Tools and Environments (CREATE) program in 2006 to develop and deploy physics-based, high-performance computing software applications for the design and analysis of military aircraft, ships, and radio frequency antenna systems (and more recently, ground vehicles) through the construction and analysis of virtual prototypes. Code development began in 2008, and eight years later, CREATE is already beginning to accomplish these goals.

Associate editor in chief Doug Post describes how the periodic table of elements is an example of big data as we know it today.

The January/February 2016 issue of this magazine presented descriptions of the US Defense Department's Computational Research and Engineering Acquisition Tools and Environments (CREATE) program and the software engineering approach for managing its programmatic risks. This article describes the software engineering methodology deployed to manage the development risks faced by CREATE, that is, the risks arising in the product development cycle and environment. The approach here is similar to the one for the management of CREATE programmatic risks and is based on a set of shared development practices. The management of these risks is especially challenging in the environment of distributed teams developing physics-based, system-of-systems, high-performance computing software anchored in the three military departments. The CREATE experience provides a concrete example of successful implementation of best software engineering practices in a computational science and engineering milieu that has historically questioned the value of traditional software engineering wisdom and has resisted the adoption of plan-centered software engineering processes. It has allowed CREATE to adopt important software engineering practices such as use case-centered requirements management, use of pilot projects to align customer and developer expectations, continuous code integration of modular components, and scalable product support models, among others.

The Department of Defense's High Performance Computing Modernization Program Computational Research and Engineering Acquisition Tools and Environments (CREATE) program is developing and deploying multiphysics high-performance computing software applications for engineers to design and make accurate performance predictions for military aircraft, ships, ground vehicles, and radio frequency antennas. When CREATE started (2007), no commonly recognized set of successful software project management practices existed for developing multiphysics, HPC engineering software. Based on lessons learned from the HPC and computational engineering communities, CREATE leadership developed and implemented a risk-based, practice-centered strategy to organize and manage the highly distributed program. This approach led to a good balance between ensuring a sufficiently structured workflow and accountability and providing the flexibility and agility necessary to create new sets of engineering tools with the capabilities needed to design next-generation weapon systems.

The guest editors of this special issue describe five articles that make up the second part of a series describing US Department of Defense software engineering efforts.

The Department of Defense (DoD) High Performance Computing Modernization Program (HPCMP) Computational Research and Engineering Acquisition Tools and Environments (CREATE) Program is developing and deploying a suite of multi-physicsbased high performance computing tools for the design and analysis of DoD ships, air vehicles, ground vehicles, and radio frequency antennas that enable DoD acquisition engineers to make accurate predictions of the physical performance of complex weapons systems (entire airplanes or ships) quickly enough to provide timely design decision data. These tools will enable acquisition engineers to identify and eliminate design defects and integration problems much earlier in the acquisition process, before manufacturing has begun, significantly decreasing acquisition time, costs, and risks while improving weapon system performance.

The high level of technological innovation required for a strong national economy and defense is only achievable with a highly productive engineering workforce. AEIC Douglass Post and consultant Richard Kendall discuss how virtual tools and processes can help.

Here, Douglass Post discusses how the use of virtual prototypes analyzed with physics-based performance prediction tools is a potential game changer for product development.

The guest editors discuss some recent advances in exascale computing, as well as remaining issues.

A new study explains why the days of obtaining performance increases due to higher processor speed are mostly over, and where we go from here.

The three papers in this special issue describe the promises and challenges of "design through analysis" product development.

The Computational Research and Engineering Acquisition Tools and Environments (CREATE) Program is charged with positively impacting the US Department of Defense (DoD) Acquisition Process via Computational Engineering for capability gaps identified by the CREATE Boards of Directors. These prioritized requirements have been characterized and are annually reconciled in terms of gaps associated with required analysis cycle-times, physical accuracy, and necessary analysis capabilities. Furthermore, the Office of the Secretary of Defense(OSD) Overarching Integrated Product Team (OIPT) has provided goals which drive the usage of CREATE software by DoD Acquisition Programs. In order to achieve these usage targets, the CREATE Program has established a governance model and thence a hosted set of services to enhance collaboration among the developer teams and the targeted Acquisition Program (AP) subject matter experts. The goals of these collaboration services are to: 1) scale and speed the feedback amongst the CREATE and AP teams, while 2) minimizing interruption to developer and user workflows. Towards these goals, the community services have been architected as discussion forums, issue tracking, and documentation. Such services are now common place in the enterprise and open-source software projects, and efficiently scale with the user community by enabling a searchable knowledge base. In addition, owing to the key capabilities of this DoD software, CREATE requires authenticated and authorized access for each member of the community. This paper summarizes the governance model and process by which the services were selected, the architecture implemented, and the challenges going forward.