This Chinese lab is researching how to ensure your driverless car or robo-surgeon does not hang
The global market for fog computing is expected to reach US$18.2 billion by 2022 as demand for devices that support this technology takes off
The last thing you want when riding in a driverless car or being operated on by a robot surgeon is for these futuristic machines to hang. Yet that could be a problem in the future if such hi-tech innovations become everyday objects, resulting in an exponential surge of data.
To tackle that problem, a group of researchers at the Shanghai Institute of Fog Computing Technology (SHIFT) – billed as the world’s first lab dedicated to the technology – are finding ways to speed up data transmission by enabling devices to communicate with each other locally, rather than route that conversation through a centralised cloud.
“Sending data from all the devices we have today to the cloud is very resource-consuming. It takes up a lot of bandwidth, which could delay data transmission,” Yang Yang, co-director of SHIFT, said in an interview. “But if we can take care of most of the requests locally by fog computing, this can reduce the burden on cloud and improve efficiency altogether.”
Established in April last year, SHIFT represents another example of China’s investments into advanced sciences including artificial intelligence and quantum computing, part of a strategic push to ensure it is not left behind in a global technological race.
Recent studies show that fog computing is poised for steady growth over the next few years. The global fog computing market is forecast to reach US$18.2 billion in 2022, up from an estimated US$1 billion this year, according to a 451 Research report.
Cloud services platforms today are not designed for the volume, variety and velocity that the Internet of Things (IoT) generates, according to Cisco, which has been credited with coming up with the term “fog computing”.
It said billions of previously unconnected devices are already generating more than two exabytes, or 2.1 billion gigabytes, of data each day. An estimated 50 billion “things” will be connected to the internet by 2020.
Offshore oil rigs generate 500 GBs of data weekly. Commercial airlines generate 10 terabytes for every 30 minutes of flight.
Yang said researchers at SHIFT are running experiments on so-called fog nodes.
Any device with computing, storage, and network connectivity can serve as a fog node. These include industrial controllers, routers and even video surveillance cameras.
Yang, who is also a board member of the OpenFog Consortium that promotes commercial standards for the technology, said the take-off of the technology could help make personal computers obsolete in 15 years.
“All we would need is a memory stick containing the software programmes we require,” Yang said. “Plug it into any machine and it turns into your ideal computer set-up, tapping on the computing power of devices around it.”
Still, researchers have plenty of challenges to overcome before fog computing can become an ubiquitous resource for everyone.
Those include ensuring that the fog nodes communicate and work together in a safe manner, and are not exploited for malicious tasks.
Yang also said a “credit history” for each device needed to be established, which would measure its capabilities based on its previously completed tasks.
For now, Yang’s group of researchers is still hard at work to make sure that different fog computing resources are interoperable in a quick and efficient manner.
“Computing resources are the fundamental basis for intelligence, for security, for decision-making. If you don’t have them, you have nothing,” Yang said.