Amazon ecosystem expands with AWS IoT: Keynote highlights | #reinvent


The second day of Amazon Web Services, Inc’s (AWS) Amazon re:Invent 2015 event in Las Vegas revealed some secrets that can’t just stay in Vegas. In a keynote address at the event, Dr. Werner Vogels, CTO of Amazon, introduced several new product offerings that were in response to the needs of the developer community.

Mulling over the 500-plus new major features and services of the past year, Vogels had a jovial look at the future with several more stunning announcements and one that might revolutionize the Internet of Things (IoT) space forever: AWS IoT, a managed cloud platform that allows connected devices to securely interact with cloud applications, as well as other devices.

Dr. Werner Vogels, CTO, Amazon 3

Constraints of the past

Vogels was quick to mention that “cloud has removed the constraints of the past,” These constraints being hardware limitations. Now we’re in a world where its software-defined hardware. With the constraints of hardware out of the way, developers are free to work on developing and innovating in their software space. This is especially possible thanks to the scalability and the infrastructure in place.

In partnership with Intel, Vogels announced the X1: two terabytes of memory that can virtualize up to 100 cores, the fastest of any of its server’s units. Intel’s Diane Bryant, SVP, general manager, Data Center Group, introduced the world to the Xeon chip from which its development was based. Brynat mentioned that it was the first of any production utilizing the Haswell line.

In addition to X1, the t2.nano was announced, which is a smaller version of the t2, boasting a whopping 512mb processor and is primarily used for developing basic applications.

Diane Bryant, SVP, General Manager, Data Center Group, Intel keynote day 2

With hardware out of the way …

There were several new product launches or expansions of existing tools within the Amazon ecosystem. Vogels announced:

Amazon Kinesis Analytics, which allows SQL-based, time-series analytics for streaming data. It works by creating a data map to describe the data coming in, adding SQL (notifications, filters, etc.), and then finally having a destination configuration output format, such as Kinesis Streams or Kinesis Firehose, depending on how you’d like to output the data.

In an effort to expand Amazon EC2 Container Service (ECS), some new features were added under the hood. The AZ-Aware Service Scheduler allows services to be scheduled for containers as needed. Docker integration with Compose was a much-applauded new addition giving Docker containers more abilities. These are all powered by the addition of Amazon EC2 Container Registry that’s been added. This registry is key in allowing the execution of proper functions as a user needs them and for them to be sent to the right location. Seemingly to top it all off, a command-line-interface was added so commands for services can be executed there.

AWS Lambda also got a few additions with Python programming language being added in addition to the current Java. As requested by consumers, VPC Support has been added, as well as long functions: those that last under 5 minutes, a scheduler and custom retry logic that can be configured.

The above announcements culminated with the revealing of AWS Mobile Hub, which is a development hub so that mobile developers cannot be restrained by the back-end complexity of mobile devices, but rather much more simply produce the cloud-based application back-end processes and easily integrate them into the phone experience with end code sources being produced as SDKs. It strives to simplify the cloud mobile development experience, and, in addition, by bringing simplicity it brings greater reliability.

AWS Mobile Hub will let you — in three steps — choose and configure your functionality, download the source application and immediately compile it for you so you just need to access it, then choose your application and where it goes.

Internet of Things – applications and use cases

While IoT is rife with examples of silly consumer electronics, there’s also considerable more depth developing. In fact, industrial applications such as with General Electric Company (GE), who places thousands of sensors on its gas turbines, shines a light on the potential of IoT. The sensor information feeds through AWS tools configured to provide GE with the actual efficiency data. Changes that are as little as one percent can have major cost consequences. There are similar situations set up in turbines, and many other industrial areas where sensors are in place to assess data.

Keynote day 2 iot platform 2

Cars could be the next platform for innovation in IoT. Dieter May, SVP Digital Business Models of BMW, explained that BMW is looking at connecting to the cloud using sensor information from all over the car, as well as continuously updating mapping technology. He sees three major trends at the moment: autonomous driving, seamless integration of the car as a powerful IoT device and mobility services gaining ground. These trends have enabled BMW to utilize IoT that’s run through AWS at the core.

Dieter May, SVP, Digital Business Models, BMW

Internet of Things – by Amazon

There’s always been a great deal of talk about IoT; however, Amazon has just made IoT even more accessible for developers. “Sensors generate data, and actuators drive the change,” according to Vogels. Here’s Vogels Model: Devices – connect to the network – security – data collection – smarts to drive the actuators.

Achieving this model in practice is quite difficult now, but AWS has created a simple solution called AWS IoT, were you can securely connect and manage devices at any scale. The security relies on X509 certificates to create individual identities for each device, and its transported through TLS. It utilizes the protocol language known as MQTT (also known as MQ Telemetry Transport) to route it. That protocol language was chosen as its fault tolerant, small and highly efficient.

The data can then be collected in any one of Amazon’s database services, including Kinesis streams for real-time reactions to be programmed. AWS IoT works by having a device gateway that acts as the MQTT protocol reader. The device gateway acts to either create reports through its database actions or drive actuators. Actuators can be driven when data rules are set up, which can be done on the device gateway as well. These actuators might translate into producing a report on the data to actually causing another physical action to take place in another motor that’s elsewhere, say in the case of a greenhouse who’s temperature is too warm and perhaps a window would be mechanically opened.

In addition, AWS Device Shadows exist, which allow users to tell devices that are offline what they should do when they’re reconnected. You can also get the last report from that device that’s in the cloud, as well as its future state, all via RESTful API.

Watch the full interview on Amazon’s Event Page, and be sure to check out more of SiliconANGLE and theCUBE’s coverage of Amazon re:Invent 2015. And join in on the conversation by CrowdChatting with theCUBE hosts.

Photo by SiliconANGLE