Lightsource BP Releases Smart-Home Energy Management App Twice as Fast on AWS

Lightsource Labs is the innovation arm of Lightsource BP, combining advanced technology development capabilities with Lightsource BP’s extensive renewable energy experience to bring smart-energy solutions to market. While Lightsource BP is focused on the funding, development, and long-term management of solar projects, Lightsource Labs researches new technologies and develops creative solutions to help households reduce their bills and carbon footprints. Its latest solution is Tribe.

Lightsource Labs began work on Tribe in a collocated data center but became interested in AWS because of the availability of managed services such as Amazon Aurora, the cloud-based MySQL and PostgreSQL-compatible relational database service. "We signed up for Aurora and it just worked," says Ioannidis. "The idea of shifting database management to AWS and not needing to hire and manage DevOps personnel gave us real peace of mind and freed us to focus on more valuable work."

The company's use of AWS steadily expanded, and Tribe was soon re-architected to run on AWS-native services. Tribe relies on Amazon Elastic Compute Cloud (Amazon EC2) for highly secure, scalable compute capacity, Amazon Simple Storage Service (Amazon S3) to store large files and documents, and Elastic Load Balancing to ensure consistent application performance.

One of the most important services that Lightsource Labs uses for Tribe is AWS Lambda, which executes serverless functions in response to events. AWS Lambda plays a key role in enabling the release velocity required for a product like Tribe. "In the digital energy space, we have to do lots of prototyping to test new policies and scenarios, and AWS is perfect for that," says Ioannidis. "Our developers and data scientists love how quickly they can test a new algorithm using AWS Lambda. They can run a script in just a few minutes, as opposed to the hours it would take to set up servers in our own data center for this. We also like that Lambda can run scripts in so many different languages, which is another timesaver for our polyglot team of programmers.”

For a product like Tribe, the reliability of AWS is especially important. "If our customers are going to trust Tribe to manage power for their entire house, we can't afford any significant downtime," says Ioannidis. "By using AWS monitoring tools and AWS Auto Scaling, we can scale up automatically if there is ever an instance that is struggling. Just as important, we can scale back down again and pay only for the extra resources when we need them—something we couldn't do in our own data center."

By helping Lightsource Labs reduce its overhead costs, AWS helps the company make Tribe even more appealing to customers. "Our business model is based on subscriptions," says Ioannidis. “AWS helps us maintain low, competitive subscription costs by consulting with us to make sure we are utilizing our infrastructure optimally and to show us where we can make improvements. That's a big win for our customers."

Another reason Lightsource Labs is all-in on AWS is the worldwide reach needed for a solution like Tribe. "Because Tribe has an in-home component that must be able to communicate bi-directionally with our cloud backend at sub-second intervals, we need our infrastructure to be hosted as close as possible to our customers," says Ioannidis. "Because AWS has global availability of cloud IoT services, we have no latency concerns about entering new markets anywhere in the world."

Lightsource Labs finds AWS services like the AWS Management Console easy to work with and understand. "The AWS Console is very intuitive, and everything is so well documented that we hardly ever need any live support," says Ioannidis. "The transparent billing and the open, honest relationship with our account team are also a big part of why we will continue to use AWS. It's really valuable for us to be able to delegate our headaches to such a reliable cloud provider so we can concentrate on our business and our customers."

As a possible next step, Lightsource Labs will evaluate using Amazon Timestream, a serverless, purpose-built database service that can collect, store, and process time series data, to expand its services. "One exciting opportunity is the possibility of helping our customers sell solar power to the grid," says Ioannidis. "To take advantage of this opportunity we'll need cloud-based AI and analytics tools that can find value in event-based data. We’re looking forward to seeing how Amazon Timestream can help us turn this possibility into a reality.”

To learn more, visit aws.amazon.com/about-aws/sustainability.