Modernizing military simulation with open systems architecture

In this Q&A with CAE USA, we discuss the SCARS (Simulator Common Architecture Requirements and Standards) program with Christopher Marshall, lead Program Manager, and Chris “SLAM” Duncan, Director of SCARS Business Development. They discuss: the current challenges in training simulation and scenarios, the benefits of virtualization and common hardware and software; and milestones and future plans for the program. SCARS is an opportunity for industry to contribute and create higher-fidelity training environments.

Breaking Defense: Tell us about the SCARS contract.

CAE’s Christopher Marshall is lead Program Manager for SCARS. (Photo courtesy of CAE).

Marshall: SCARS is a 10-year ID/IQ to develop a centralized enterprise solution that establishes a common open-systems architecture for U.S. Air Force (USAF) simulators, platforms, and the Joint Synthetic Environment. Through a single USAF operations center and shared infrastructure backbone, SCARS allows all USAF simulators to leverage common applications, and rapidly update and maintain evergreen modeling and simulation capabilities.

CAE acts as a facilitator for the Air Force training vision. We host and organize the SCARS Engineering Capabilities Board and work with MAJCOMs and industry partners like Leidos, CymSTAR, and Dell to collaborate and generate a reference architecture that makes sense and enables all these training systems to run on the same common SCARS OPE (on-premise equipment).

As we move forward with more joint and distributed training, this solution provides common training system architectures utilizing common applications, allowing us to accomplish the actual training and tactics development we’re looking for at considerable cost savings and higher fidelity.

Breaking Defense: What’s the ultimate goal of the program?

Marshall: The vision of SCARS is to create a cyber-resilient posture and provide the ability to utilize common models across all training systems – to train and test virtually, using the same data with an enterprise-level solution. The foundation of this vision is dependent on fully virtualizing the training systems onto a common hardware stack with a common architecture – the SCARS OPE.

Once the system is virtualized onto that hardware stack, you achieve an advanced position to gain efficiencies with cyber patch, STIG, remote scanning, and utilization of common applications that significantly increases interoperability between existing disparate training systems.

Chris “SLAM” Duncan is Director of SCARS Business Development. (Photo courtesy of CAE).

Duncan: The SCARS program creates a paradigm shift in traditional modeling and simulation support for the USAF. In the past, simulator builders have built products that are highly dependent upon proprietary tech stacks stove-piped into only supporting a single weapon system. By utilizing SCARS, a marketplace to create high-quality content is now available to support every platform and warfighter.

This has been an objective for USAF simulation training for quite a while and, because of the vision provided by Col. Matt Ryan’s team we’re now on the cusp of achieving. We’ve made some tremendous progress with the program recently in large part due to USAF leadership and focused efforts.

CAE’s role in this is to execute the USAF vision without introducing more proprietary challenges. We’re providing an enabling tech-stack solution and connectivity to integrate Air Force weapons systems and provide content to the warfighter.

Marshall: The SCARS program is chartered to not only work with government and industry to develop a common training system architecture and source common models but also establish configuration control for all users within the SCARS library. It’s imperative that the source data be configured, controlled, and distributed from a centralized source for all enterprise use to flush out interoperability issues and create a fair fight environment so we can trust the training and also the data needed for analysis. We must reach a point where we can trust the virtual data and results for live fly comparison and tactical development.

Industry plays a big part in achieving this vision through high-fidelity models. Today, a significant amount of funding is required to develop and maintain over 50 different models for the same capability (radar, weapons, Link 16, etc.) at various levels of fidelity for the USAF. The SCARS vision enables best-of-breed models for training and simulation for better results and overall government cost savings. Those cost savings can be applied to higher fidelity development and innovation. That’s what the government needs.

Breaking Defense: What is SCARS replacing?

Duncan: SCARS isn’t replacing anything outright; it’s a response to the Air Force’s challenge of training its combat forces in a constrained environment, where access to consistent data across simulations has been limited. The existing simulator landscape comprises custom solutions from various contractors and standards, resulting in disparities in fidelity and compatibility.

Marshall: Custom training solutions are built by 30 different contractors and over 50 different platforms within the Air Force alone. They have various levels of fidelity and training due to age and differing requirements.

If you try to plug them all together, they don’t work with each other. Their terrains are different; their imagery, their weapons, their models, nothing is the same.

SCARS is a response to the Air Force’s challenge of training its combat forces in a constrained environment, where access to consistent data across simulations has been limited. (Image courtesy of CAE)

Breaking Defense: Is JADC2 and all-domain operations driving the need for access to the same data across different platforms?

Duncan: While JADC2 and all-domain operations play a role, the primary driver behind SCARS is the increasing importance of simulators in training to modern warfare scenarios, which demand access to realistic data, electronic warfare simulations, and network-centric training experiences not easily achievable in live training.

Establishing a common simulation training architecture across the Air Force isn’t a new concept. It’s been attempted before. I believe the difference this time is the success of the Joint Simulation Environment, technology maturation, and the urgency to deliver a high-fidelity solution to an increasingly complex training requirement for our airmen.

In addition, the USAF will realize a tremendous cost savings in some of the more benign requirements like hardware, and instead provide more incentive to industry for developing models and applications. For example, instead of supporting 50 different AMRAAM models that are similar but support separate weapons systems, there might be just one for each missile variant that gets shared across the enterprise. That concept applies to many different services like weather, electromagnetic operating environment and sensors, threats, datalinks, etc.

Breaking Defense: A 10-year ID/IQ is a long time. What are the milestones involved in creating this full virtualization?

Marshall: First and foremost, we needed to show that we could accomplish the cyber task – higher frequency in scanning and higher frequency in patch/STIG. We do that across 18 different sites right now. For the remote patch and STIG, we’ve completed enterprise testing at three sites and two different platforms.

We are accelerating full virtualization, a full SCARS vision on the F-16 platform. By accelerating F-16, we not only have a proof of concept for a cyber-resilient training solution and common model integration, but it also provides the metrics concerning duration, cost and benefits.

Duncan: Once we better understand the success of virtualizing on F-16, we can expand the envelope. This is where it gets interesting. Because of SCARS, the USAF will be able to support the growth of the Joint Simulation Environment to a wider audience.

Breaking Defense: What other benefits are there to virtualizing onto the SCARS OPE?

Marshall: We’re seeing significant benefits related to the notification earlier this year that Microsoft will no longer support Windows 10. The government issued a Rough Order of Magnitude request to platform contractors to estimate what it will take to go to Windows 11 by the end of 2024.

We worked with the Air Force Simulators Division to start collecting costs on what the government anticipates the impact to be. We stopped counting when we reached over $500 million, a considerable government bill. And it’s something that comes every three to five years.

Since it’s unknown, it becomes an unfunded request, meaning that the budget to become compliant and avoid cyber vulnerability will come out of development capability or sustainment budgets. A significant portion of that cost, over $500 million, as mentioned, is in the hardware requirements for Windows 11.

The SCARS OPE servers being deployed are already compliant with the hardware requirement for Windows 11. We’ve started putting together some documentation and are doing a proof of concept on F-16 to collect metrics on what it will cost to virtualize the Windows components onto the SCARS OPE.

We already know that leveraging the SCARS OPE will be a fraction of the cost of transitioning from Windows 10 to 11 for the F-16 program. Still, the most significant impact is future migration when the next End of Life notification is released. In the future, when the Windows 11 notification comes out for the move to Windows 12, we’ll be poised to accomplish that for F-16 with pennies on the dollar, as it’s far more efficient to upgrade a virtual OS. In addition, common models are deployed as highly portable containers that are platform agnostic.

Finally, all SCARS OPEs are supported from a SCARS Operations Center (SOC) by classification level. This connection from a central operations point of service is facilitated by leveraging existing partner networks or the SCARS Wide Area Network. The SOC supports not only remote cyber scanning capability and patch/STIG deployment but also delivery mechanisms for common models or training tools.

A training system, virtualized and rehosted onto a SCARS OPE, affords significant enterprise benefits. Once rehosted to the OPE, patch/STIGs can be applied to the training system directly from the SOC after being verified and approved. Affiliate and platform training system updates can be simultaneously pushed from the SOC across all the platform’s training systems, such as maps, OFPs, threats, weather, etc. That is critical not only for maintaining common configuration but also for concurrency and cyber compliance. That is the vision, the goal, and the means to train as you fight for the USAF.