How Amazon's cloud will make the IoT smarter
AWS IoT platform could upgrade connected devices over decades to keep them up to date
Billions of connected devices in the home, office and industry will soon be able to hook into Amazon Web Services (AWS) to link up to other devices and grow smarter over time.
The tech giant today announced the availability of a beta version of its Internet of Things platform, AWS IoT, which will be able to power and support billions of remote sensors both online and offline, encrypting communications between them.
The platform will allow customers to store, process, analyse and base business decisions on IoT data they put into it, while manufacturers can automate actions for devices to perform when certain data is received - such as sending alerts when a motion sensor is triggered.
Citing the streaming video kit Dropcam (now owned by Nest), Amazon CTO Werner Vogels pointed out the extent of what now counts as IoT data, saying: "Video is no longer just something to be watched, it's something to be analysed."
But Amazon is also tackling a major IoT issue - that of patchy coverage, where poor internet connections could stop devices working.
AWS IoT is able to create a virtual version (Amazon calls it a 'shadow') of each connected device that includes details of the device's status, and remains always available, so any actions an app sends to that device take place as soon as a connection is re-established.
What's more, the IoT platform should make smart devices smarter by leveraging the cloud to do the heavy lifting, while the devices will likely be as cheap and lightweight as possible as manufacturers try to keep their costs down.
Get the ITPro. daily newsletter
Receive our latest news, industry updates, featured resources and more. Sign up today to receive our FREE report on AI cyber crime & security - newly updated for 2024.
AWS's product strategy head, Matt Woods, told IT Pro: "If they are going to be as cheap and efficient as possible we can't really rely on them to have much computational power, but you do want the aggregate system to be as smart as possible.
"By housing it in the cloud, by providing machine learning techniques right next to it, those devices are going to get really smart really quickly."
He added that another goal is to keep the devices very up to date via software updates delivered via AWS IoT, despite them being in service for decades.
"You'll be able to continue to add features over the next 10, 20 years. That's one of the goals, is looking at it less just like one device talking to the cloud, but in aggregate the graph devices - how do you orchestrate and drive as much intelligence and smarts for the devices they are actually connected to at the other end of the cloud?"