For Automakers, the Road Ahead for Innovation Runs Through the Cloud

Learn the benefits of a cloud-native approach to software development in the automotive industry.

September 27, 2022

Traditional automakers face unprecedented upheaval in today’s competitive automotive market. As such, a cloud-native approach to software development is critical for accelerating the industry’s digital transformation. Yu Fang, CTO and founder, Sonatus, discusses the key benefits of shifting to this development.

Traditional automakers face unprecedented upheaval, with supply chain shortages, a shift to electric vehicles, and searing competition from new entrants. But we see a bright future ahead if they take the lead from what has happened to software development since the early 2000s.

Over the last few decades, automakers have increasingly turned to their suppliers to build and design the components and subsystems that makeup automobiles. The same pattern continued as electronics took over more and more in-vehicle systems.

The result is that modern automobiles feature dozens of bespoke Electronic Control Units (ECUs) from numerous vendors, based on disparate silicon, running software that is hardware-specific and difficult to update or reuse. Standardization of both software and hardware is minimal, even as the number of lines of code rises above 100 million in a high-end vehicle.

The result is a system that is complex, inflexible, and “out of gas.” It is the reason why software-related defects account for more than halfOpens a new window of all vehicle recalls, according to Stout’s Automotive Defect and Recall Report, 2020. And it is why newcomers like Tesla, whose vehicles incorporate modern software and hardware architectures, are pulling ahead. So, automakers have realized they need to re-architect not just the software and underlying hardware that controls their vehicles but also how they produce them.

In many ways, these challenges are not surprising, particularly for those with a background in the enterprise tech world. That is because, over the last couple of decades, the tech industry has had to embark on its own journey towards standardization and interoperability.

So can enterprise tech plot a route forward for vehicle manufacturers? We believe it can.

People often describe a modern automobile as a smartphone on wheels. But perhaps a better description is a data center on wheels, with scores of computers and other devices communicating and interacting in complex ways over high-speed networks. And when it comes to data centers, the state of the art is the software-defined model which underpins today’s most successful cloud-based businesses and services.

Being software-defined means functionality is overwhelmingly implemented in software, which is abstracted from the underlying hardware, making it easier to update and reuse. As importantly, it also means the software is designed to be dynamically configured and programmatically controlled by other software, enabling capabilities like real-time optimization, automated diagnostics, and proactive security.

See More: Why Leading Organizations Are Investing in Dedicated Devices

Driving While Abstracted?

Increasingly, vehicle software is being designed using cloud native technologies and architectures, such as microservices and containerization. Applications and individual functions and features are composed of modular, portable services that interact via APIs. Fixes, upgrades, and even new functions and features are delivered as incremental updates.

Cloud-native development builds on this by incorporating development and vehicle simulation environments, allowing vehicle software to be developed, tested, and verified completely in the cloud rather than requiring hardware-based test environments. Validated updates and changes can then be deployed remotely and at scale. In the case of a software-defined data center, this removes the need for a technician to update individual pieces of hardware on-site. In the case of a software-defined vehicle, it removes the need for vehicle owners to head to a workshop for a critical update.

So what other benefits can we realize if we can apply the cloud-native approach in the automotive sector?

Currently, much of the industry focus is on how software-defined vehicles can receive bug fixes over the air. This is indeed a pressing need, given the upheaval caused by defects in existing software. Closely associated with this is the ability to deliver new features and services over the vehicle’s life to increase customer satisfaction and loyalty and enable new profit streams.

But we believe the benefits of cloud-native development in the auto industry go far beyond this.

First of all, cloud-native development can accelerate innovation for vehicle manufacturers. By breaking down vehicle software into modular, reusable, updatable services, development teams no longer have to reinvent the wheel whenever they need to update an individual component. And when they develop a new feature, they can focus on that service or its container, knowing other services in other containers will continue to function without changes. Further, testing software components becomes less burdensome as this can be done in the cloud using a vehicle simulation environment rather than requiring an actual vehicle.

Additionally, decoupling software from hardware allows software development to accelerate ahead of slower hardware timelines. It also gives manufacturers more flexibility when choosing silicon and makes them less vulnerable to chip shortages.

See More: The IoT Meets Process Control: Strategies for the Future

A Really Agile Vehicle

A cloud-native approach also enables a more dynamic vehicle. As individual components and subsystems are better connected with common APIs and protocols, they become more configurable and programmatically controllable. This allows software in the vehicle, or the automaker’s data center, to analyze vehicle data, identify issues and optimizations, and even automatically enhance the vehicle without requiring new software. Put simply, the system can see what is happening, make sense of it, and] take action to address it.

In summary, the cloud-native approach is simply more efficient and agile and ultimately speeds up innovation. When an automaker implements a software-defined platform at the heart of its vehicles and establishes a cloud-native development environment, it can react far more quickly to new opportunities, issues, regulations, and risks.

What might this all look like in practice? The recent changes to E10 gasoline regulations in the United Kingdom provide a good example. If software-defined vehicles needed tweaks to accommodate new fuel formulas, those changes could be delivered from the cloud rather than necessitating a trip to a dealership. And automakers and fleet managers could get real-time data on the precise effects of the fuel changes and of the updates they have deployed, allowing them to optimize performance further.

There is still some way to get there. But industry collaboration efforts like the Scalable Open Architecture for Embedded Edge project [SOAFEE] are bringing automakers together with silicon makers, software developers, and cloud providers to make cloud-native development a reality for automotive. Together, we are confident that such efforts will deliver far more than easier updates. Ultimately, they will help enable the wholesale digital transformation of the auto industry, resulting in better, safer, more secure, and always current vehicles.

The future of the automobile may be unwritten, but we are pretty sure it will be software-defined and cloud-native.

As an automaker, have you taken a cloud-native approach to software development? Let us know what benefits you have seen on FacebookOpens a new window , TwitterOpens a new window , and LinkedInOpens a new window .

MORE ON TRANSFORMING TECH FOR AUTOMAKERS

Yu Fang
A seasoned entrepreneur, technologist, and executive, Yu brings over 20 years of experience leading high-performance teams and building cutting-edge technologies and products for the data center, eCommerce, and automotive markets. Before founding Sonatus, as Head of Mobile Engineering, Yu led a team that built AI-based software on mobile platforms to improve the driving safety of commercial fleets. Before that, as SVP of Engineering at Tanjarine and CTO at Cardfree, Yu led the development of a mobile and cloud-based eCommerce platform serving tens of thousands of stores and tens of millions of customers.
Take me to Community
Do you still have questions? Head over to the Spiceworks Community to find answers.