The new AWS services include a service to get started in IoT, one to manage fleets of devices, ones for security and analytics, one for small devices without CPUs, and one to run machine learning at the edge.
Andy Jassy onstage
LAS VEGAS – Using the AWS clouds to capitalize on the developing mass growth stage of the Internet of Things has been a major theme at the AWS re:Invent event here. To capitalize on that growth – and overcome barriers to adoption – AWS announced six new IoT services on Wednesday. They are AWS IoT 1-Click, AWS IoT Device Management, AWS IoT Device Defender, AWS IoT Analytics, Amazon FreeRTOS, and AWS Greengrass ML Inference.
“We are now in the next phase of the Internet of Things, where the growth of devices will be exponential,” said AWS CEO Andy Jassy in his Wednesday keynote. “The first issue though is just getting in the game.”
AWS IoT 1-Click was created to make getting into the game easier, enabling devices to perform simple functions like single-button devices that do things like call technical support, reorder goods and services, or track asset locations
“AWS IoT 1-Click allows for one-click creation of an AWS Lambda trigger for any device,” Jassy said. All the customer has to do is download the mobile app, registering and selecting an AWS IoT 1-Click enabled device, and associating an AWS Lambda function with a single click.
“You can choose from prebuilt Lambda functions or author and port your own,” Jassy said. “This will help a lot more companies to get into the Internet of Things.
AWS IoT 1-Click is now available in preview.
Jassy then indicated how AWS has addressed another issue – that management of the vast network of devices, from different vendors, has been a pain.
“Device management now is still time consuming and hard, and there have been no end-to-end solutions stitching all the third party solutions together,” he stated. AWS IoT Device Management lets customers securely onboard, organize, monitor, and remotely manage IoT devices at scale throughout their entire lifecycle.
“AWS IoT Device Management securely onboards, organizes and manages devices at scale,” Jassy said. “You can provision an entire fleet with one click. You can then decide to manage the fleet as a single entity, or in parts, or even just manage one device.”
AWS IoT Device Management is available now.
The new security service announced, AWS IoT Device Defender, is the most distant service announced, with availability slated for some time in the first half of 2018.
“IoT is still a relatively new medium, and connected devices are a weak spot,” Jassy said. “Our IoT Core Service is great for a single device. AWS IoT Device Defender is designed for fleets of devices.”
The service lets the user define expected behavior, like which ports should be open on a device, where the device should connect from, and how much data the device should send or receive.
“It then lets you audit policies, monitor devices for abnormal behavior, and alerts you if something is off,” Jassy said. This ability to identify anomalies and generate alerts is the heart of the service.
“It will be a big step forward in handling the security of all devices,” Jassy stated.
“A new AWS IoT Analytics is for customers who want basic analytics for connected devices,” Jassy said. “It is a fully managed analytics service that cleans, stores processes and analyzes device fleet data.”
Customers only have to identify the device data they want to analyze, and they can optionally choose to enrich the device data with IoT-specific metadata like device type and location, by using the AWS IoT Device Registry and other public data sources. AWS IoT Analytics also has features for more sophisticated analytics, like statistical inference, and can get more by using Amazon QuickSight in conjunction with IoT Analytics.
Jassy then turned to another persistent management problem with the IoT, the fact that the vast majority of devices that can connect to the cloud don’t have a CPU, because they aren’t large enough. Instead they have an MCU, a microcontroller, and so can’t take advantage of existing AWS IoT services.
“These smaller devices with an MCU outnumber ones with a CPU by about 40-1, because only devices that are big enough or expensive enough house a CPU,” he stated. “So the vast majority of devices aren’t connected to the cloud and cant leverage it. That was a problem we had to fix.”
AWS’ answer is Amazon Free RtOS. FreeRTOS itself is an open source OS for microcontrollers that lets them perform simple tasks, but which on its own, doesn’t address the issue because it wasn’t designed for IoT.
“We already have this array of supported microcontrollers, so we are extending FreeRTOS with Amazon FreeRTOS, which contains software libraries that let customers connect these small devices to AWS cloud services like AWS IoT Core or AWS Greengrass,” Jassy said. “This makes it easy to connect these devices to the cloud, and leverage the cloud like CPU units.”
The final IoT announcement is a new feature for AWS Greengrass that brings machine learning to the edge.
AWS Greengrass ML Inference is a new feature of AWS Greengrass that lets application developers add machine learning to their devices, without requiring special machine learning skills.
“Developers can build and train models in the cloud, use Greengrass to transfer models to their devices, which can then spit them out to production environments at the edge,” Jassy said. “We expect this will create a huge increase in the ability to run machine learning at the edge.”
