Elevating cluster software development with QNX Hypervisor on AWS

Digital cockpit systems are the primary “vehicle to user” interface of a Software-Defined Vehicle (SDV) which include digital instrument clusters that have sophisticated graphics needs as well as safety-critical requirements. For instance, vehicle state information, such as vehicle malfunction warnings, gear position, and advanced driver assistance state are among the features that must be available to drivers at all times to help ensure road safety. According to a report from Boston Consulting Group (BCG), “The emergence of SDVs will create over $650 billion in value for the auto industry by 2030, making up 15% to 20% of automotive value.” This highlights the massive opportunity for automakers who will have to cultivate SDV capabilities to manage technical complexity via cross tier development, connection to regional and global supply bases, and efficiently integrate platforms.

To help unlock this opportunity and expedite the development of digital instrument cluster software, automotive Original Equipment Manufacturers (OEMs) and their suppliers should be able to build and test software without the dependence on physical hardware systems. Heavy dependency on specialized hardware target systems, that are often expensive and scarce, has been a hurdle for OEMs and has kept them from launching new vehicle features rapidly. BlackBerry QNX Hypervisor helps solve this challenge by reducing an OEM’s dependence on hardware and enables OEMs in launching cloud targets with binary parity. The QNX Hypervisor is a foundational component designed to help establish stability and integrity of mixed criticality systems and safety functions. In this blog post, we delve into how the QNX Hypervisor technology, together with Amazon Web Services (AWS), can help enhance the development and testing of such a consolidated digital instrument cluster domain software. We will discuss how decoupling software development from dedicated hardware targets can increase software development efficiency.

Digital cockpit and cluster software development with QNX
As shown in Figure 1 below, an automotive hypervisor offers virtualization technology that helps enables the secure separation of multiple virtual machines. The hypervisor manages the virtual machines and also controls how they communicate with each other. Each virtual machine can run its own operating system (OS), known as a guest OS, and applications.

The QNX Hypervisor isolates resources used into safety critical and non-safety critical vehicle functions and protects against interference between the operating environments (e.g. virtual machines) in the system. Freedom from interference (FFI) is a cornerstone capability for building mixed-criticality systems. OEMs are consolidating different software domains with varying Automotive Safety Integrity Levels (ISO 26262) into a single cockpit domain controller as illustrated in Figure 2 below:

BlackBerry QNX can provide virtual peripheral devices that require simulation or emulation per VIRTIO standard (to the guest OS) allowing utilization of the same application interface abstraction between cloud and in-vehicle software. By following the VIRTIO standards for supporting device sharing between guest and hypervisor, the guest environment remains the same when running on QNX Hypervisor in the cloud or on a QNX Hypervisor in an embedded target. This opens up the opportunity for a wide range of scenarios, that can be integrated and tested using software-in-the-loop methods, before final verification and validation on the actual hardware target environments.

Why digital cockpit and cluster software development is easier with QNX Hypervisor on AWS?
AWS coupled with the QNX Hypervisor offers scalability for developing and testing automotive software for mixed criticality software with varying testing scenario demands. Safety-critical software development by OEMs demands additional test methods like fault injection and resource usage tests among many other requirements and processes. AWS provides the ability for customers to allocate their computing resources dynamically; this scalability helps the development process remain responsive to evolving requirements.

Automotive manufacturers can harness the power of Amazon Elastic Compute Cloud (Amazon EC2) and QNX Hypervisor together to help accelerate SDV development. Amazon EC2 instances powered by AWS Graviton2 processors offer access to powerful graphics and compute environments. This enables OEMs to complete development and validation of software, which is deployed on heterogeneous computing environments. This enables development and testing experiences with both binary parity and performance parity with the actual cockpit domain controller. For Amazon EC2 instances without GPUs, BlackBerry QNX also provides software-based GPU and audio to enable development of graphical and audio-based applications such as those for In Vehicle Infotainment (IVI) and cluster. Additionally, the aarch64 instruction set architecture parity between in-car and in-cloud processors, allows developers to execute the same application images (binaries) on both computing environments.

The QNX Hypervisor on AWS also facilitates remote collaboration between globally distributed teams. With the global reach of AWS, a homogeneous development and target system runtime environment is accessible from anywhere in the world. This is a great advantage because most OEMs and their suppliers develop vehicle software functions that target many different geographies. The QNX Hypervisor helps enable the scalable development of software incorporating industry specific regulations and safety requirements. In addition, the availability of QNX Hypervisor on AWS allows OEMs easy adoption of modern software development methods like agile framework, continuous integration and continuous delivery (CI/CD), test driven development and many more.

Finally, the availability of QNX Hypervisor on AWS allows customers to use AWS’s pay-as-you-go pricing model. It eliminates the need for large upfront hardware investments and makes it a more cost-effective solution for automotive manufacturers. For monitoring, tracking, and control of the project budgets. AWS provides services like AWS Cost and Usage Reports for cost-tracking and reporting and AWS Budgets for defining cost budgets and thresholds.

Conclusion
By using the QNX Hypervisor on AWS, automotive OEMs and their suppliers can now accelerate their development and validation of their safety-critical features, identify defects earlier in the development lifecycle, and reduce both time to market and cost of development. With cloud scalability, a performant validation environment, global collaboration, and robust security, AWS and BlackBerry QNX are helping empower automotive innovation.

The QNX Hypervisor Amazon Machine Image (AMI) will be made available globally for early access through AWS Marketplace. If interested in exploring how this AMI can help accelerate your automotive software development, you are now able to request early access to QNX hypervisor via the QNX Website.