Data Backup Digest

You know about data storage in personal computers, but data is booming all over the world in appliances and technology that might not first come to mind. Refrigerators, lighting systems and alarms are all things that can now require ample data storage thanks to smart homes and device connectivity.

The same applies for cars, which nowadays hold large amounts of storage thanks to all of the safety features and tech that is being included. Of course, adding more sensors to make the vehicle more secure isn’t a bad thing, but having these produces more data, which then has to be processed and stored.

This year, a new car with 2 cameras, 16 sensors and 1 to 2 hours of daily drive time will produce around 20GB of data each day. Next generation cars will include over 20 computers, at least half of these requiring data storage. This is quickly creating a data problem – how is all of this going to be stored?

It’s currently usual for each computer in the car to have its own storage. These can be things like event recording for self-driving cars, in-vehicle infotainment, advance driver assistant systems and more. It’s not new for cars to need storage; they’ve had microprocessor control units, boot ROMs and program flash controllers for years and that isn’t going to change.

To cater for the infotainment systems found in cars, DDR storage has been used to account for the graphic processors. For example, some systems need storage to store song libraries. Nowadays it’s more common for SSDs to be used due to the high vibration environment in cars, which a standard HDD doesn’t cope well with due to its spinning disks.

The increase of Advanced Driver Assistance Systems (ADAS) in cars means that there’s a much higher-performance processing requirement, which in turn impacts the memory architectures.

“64-bit processors in a car is new, and that started with a lot of the ADAS vision processors that need 64-bit operation. These also need a vision co-processor, like an embedded vision co-processor sitting next to the host processor, and this is also running at pretty fast data rates with very advanced vision algorithms,” says Ron DiGiuseppe, senior strategic manager for Synopsys Solutions. “When you throw in the requirement of vision co-processing with 64-bit host processing, the external memory requirements are growing in terms of density and performance.”

It's unclear precisely how much data will need to be stored in cars in the future – nor exactly what type of data it’ll be. When autonomous vehicles were a concept, it was imagined that much of the processing would be handled in the cloud, and that they’d hit the market in 2030. With Ford now announcing that they’ll have them on the road in the next four years, and Telsa already including them in their cars, it’s clear that technology is speeding along far faster than predicted.

As such, that’s put pressure on storage to keep up. It’s no doubt that data storage will have a big role to play in the cars of the future; as will the processing architectures behind them. The data will need to be moved, processed and stored quickly and in great amounts; some will need to be recalled quickly, others stored for longer periods of time. And all of this needs to be happening in a system that puts safety as a critical factor.

It's a difficult design to cater for, made more difficult by the drastically changing car market, and it’ll be interesting to see if storage technology can cater for it.