- June 30, 2017 /
- by Tiffiny Rossi
The connected car easily conjures up topics like 5G, V2X, and the cloud. It’s certainly enchanting – all the amazing services the cloud and high-speed wireless connectivity will bring to our cars in the coming years. But the down-to-earth reality is the mission-critical handling and processing of data will be done inside the car itself.
The noble edge may lack glamor, but certainly not relevance
Opposite of cloud computing, data processing and analysis that happens near where the data is collected is called “edge computing.” It’s hard to get people excited about edge computing – it’s just what we’re used to. Moreover, only a few years ago the jury was still out about the importance of edge computing in the IoT era. Today, its role is strong and clear, especially when it comes to the connected car. As venture capitalist Peter Levine explained to CIO Magazine, “the driverless car, whose 200-plus CPUs effectively make it a “data center on wheels,” is a prime example of an edge device whose computing capabilities must be self-contained.”
Levine further explained that an autonomous vehicle that relied too heavily on data from the cloud could run through stop signs and crash because of latency issues. And, as SanDisk’s Martin Booth reminds us, a connected car is not truly connected at all times, nor is the connection always optimal. While latency issues are expected to decrease over the coming years – especially with advancements in 5G technology – experts don’t feel it’s feasible to push all connected-car data to the cloud.
This point was driven home at TU-Automotive Detroit 2017 by Dan Gittleman, CEO of Xevo – a company specialized in data-driven automotive solutions. At his event talk, Gittleman said, “Think about it, there are hundreds of millions of cars and each car can have hundreds of sensors in there. It’s just not possible to get all this data to the cloud.” And, according to Gartner, “all this data” may reach over 280 million gigabytes annually by 2020.
Gittleman added that a lot of this data needs to be processed on the cars themselves – at the edge. There, the system can decide what data needs to be pushed to the cloud, what data needs to stay in the car for immediate action, and even what data can be tossed away entirely.
Connected cars need fast local storage solutions
On a whole, connected car data, and the business models emerging from it, could be worth $1.5 trillion by 2030 according to a 2016 report from McKinsey&Company. This is a valuable commodity to preserve – which places greater focus on onboard data handling and storage solutions. According to SanDisk’s Martin Booth, advanced autonomous vehicles may soon need edge storage capacities of one terabyte or higher.
Advancements in flash memory technology allow us to record the massive amounts of data pouring into the connected car at very fast speeds. But that also requires ultra-fast data storage software that can reliably handle multitudes of simultaneous data streams. Together, the storage hardware and software work to preserve the integrity of all the valuable data moving through and stored inside the connected car.
While all eyes are turned toward the cloud, the not-so-lofty reality is that local means of handling, processing, and storing data in connected cars are not losing relevance. Into the foreseeable future, technologies that support edge computing, including onboard sensors and edge storage solutions, are only going to play an increasing role. And for us at Tuxera, we’re happy to be right there – at the edge, close to the data.
Attending TU-Automotive Europe?