Architecture and Coding Guides

---

Overview

Some of the major challenges with DoD software programs are the speed of execution relative to an approved schedule and an inability to scale the software products to meet the needs of the Warfighter. This applies to both systems that require mission-critical elements (safety, security, performance) and Science and Technology programs that are intended to create transitional knowledge and resources for the Program Management Offices. As noted in the Defense Innovation Board’s (DIB) report on ‘Do’s and Don’ts for Software’, programs are divided into development and sustainment limiting the program managers from adding new capabilities later in the lifecycle of the program when change is needed most. Additionally, the sustainment teams may not have the development competencies that created the original software system and only provide point solutions, often at great expense. This is compounded by the need for supportable and upgradeable software systems delivered at the pace of mission needs.

Architecture Design Process

Architecture Centric Virtual Integration Process is a good starting point to capture and communicate the structure, boundaries, constraints, and attributes but this must also be testable, scoped, and validated along with any other components in the overall system (and it relates to a broader ecosystem). Key elements of the architecture must include validated requirements with the understanding of performance and quality needed to address the attributes of the architecture from a static perspective while addressing the lifecycle of requirements change from development through deployment.  This allows the architecture to transform independent of the components that support the functional behaviors that are intended to be sustainable, stable, and adaptable to changes that will come with new technologies, mission requirements, user needs, and developer workflows. 

Modeling and Visualization

Model-based Systems Engineering (MBSE) provides the platform for visualization, validation, communication, and collaboration in a modern digital engineering environment. These tools, however, fall short in their ability to predict the performance and behavioral nuances of a functioning software system and must take into account the incremental tests, prototypes, and proof points to develop trust and confidence in the design in parallel with the architecture modeling activities.  With an incremental and iterative approach to get the baseline architecture into an operational and testable state at a high level and a detailed viewpoint.

Important Considerations in Software Development

• Minimal Standards Alignment
• Documented Naming Convention
• Clean Core Architecture with Component Boundaries
• Well Defined Configuration and Threading System
• Abstract Systems and Services
• Utilities and Dependencies
• Integrate Debugging and Exception Handling
• SOLID in Practice
• Object Calisthenics
• Clean Code