EnergyAlert is a service provider focused on monitoring and analyses of energy consumption. Its’ solutions cover a wide range of energy sources and provide insights not only on how much energy a specific tenant consumes, but also on how to best leverage the granular information.
As an early adopter of the AWS public cloud, EnergyAlert understands the competitive advantages services like Amazon EC2 and Amazon RDS MySQL can provide. But to deliver a disruptive solution, they needed to go further and make use of every available resource.
The previous energy monitoring application was already running on AWS. It basically received data from the metering companies via email and web services. But energy monitoring is changing. From monitoring buildings via their point of connection to the grid, now, metering is on a more granular level like rooms or devices.
To make this kind of reading feasible, EnergyAlert knew they would need to use IoT services and techniques to allow monitoring on a large scale. And however the solution to be would be created, it had to account for 3 things:
- The IoT devices used should not be dependent on the internet connection of the customer to transfer their data to EnergyAlert.
- It had to allow devices to run on batteries for an extended period of time (minimum 1 year)
- The system should cope with devices on a large scale (100.000+)
Oblivion proposed the use of the LoraWAN technology to allow IoT devices to send data to EnergyAlert without requiring an internet connection. LoraWAN is a low power, long range, communications solution. It uses centralized gateways that receive messages from IoT devices and send those messages to a router.
A LoraWAN-stack on top of Amazon EC2 will lookup applications and gateways in Amazon DynamoDB and deliver messages to the AWS IoT service using MQTT. AWS IoT handles distribution systems for archiving using Kinesis Firehose in combination with Amazon S3 and processing with AWS Lambda, which will transform the data and post it to the EnergyAlert application stack.
All the data is securely stored in Amazon S3 and Amazon Glacier with appropriate lifecycle policies that will ensure at the same time availability, durability and cost optimization.
The new solution for EnergyAlert is a success. It grants granular monitoring to a massive amount of IoT devices. All requirements imposed were met, and the architectural design accounts for AWS best practices. As a result of the used AWS services (AWS IoT, AWS Kinesis Firehose, Amazon S3 and AWS Lambda just to name a few), the solution has proven to be highly available, scalable, cost effective and reliable.
Internet of Things (IoT) and serverless architecture: how our clients use data of massive amounts of IoT devices.