OST Helps Traeger Grills Migrate Over 110K Devices to AWS IoT with Zero Downtime

Traeger Grills, a household name for smokers, grills, and barbecue products, launched a popular line of Wi-Fi-enabled pellet grills in 2018 that operated via an Internet of Things (IoT) platform. However, when the IoT vendor decided to sunset its platform, Traeger faced a time crunch to migrate customers and rebuild its solution, without disruption to current customers. With the help of Amazon Web Services (AWS) IoT Competency Partner, OST, Traeger migrated 111K devices and 700K mobile users to AWS IoT Core in just three months.

Smoking meat normally requires painstaking focus and hours spent in front of a grill. But with an IoT-connected Traeger Grill, customers can control the entire smoking process through an Android or iOS application. By making it easier to smoke and grill food, Traeger unlocked new adoption and soon saw major traction for its smart grills. But just as the products began scaling to take on new users, the IoT platform vendor decided to shut down. Traeger was forced to migrate its customers and quickly launch a new IoT platform. “We did the market research and saw that AWS IoT was the best place for us to migrate in the short term, but also gave us the most long-term advantages to grow past what our leadership and investors thought possible,” said Ryan Tarver, Director of Cloud, IoT, and Data at Traeger Grills. “We didn’t have the capabilities internally to handle the migration, so we worked with OST, who has been a phenomenal partner.” To maintain the connection of each device during the migration, OST stood up the new IoT platform while the former one was still operational, so they could exist in parallel.

OST first developed a new IoT platform on AWS using OST’s IoT Foundation and AWS IoT Core as the basis for connected grills to interact with the end-user application. In addition, it used AWS IoT Device Management so Traeger could track, monitor, and manage its fleet. FreeRTOS provided the real-time operating system for a secure connection to the AWS Cloud. From there, OST customized the experience, leveraging Amazon Alexa Voice Service Integration for AWS IoT Core for voice capabilities and creating a personality for devices. Integrating the platform into existing device hardware and firmware laid the groundwork for the migration. To carry out phase two, OST built a serverless data lake that runs AWS Lambda for compute and uses Amazon Simple Storage Service (Amazon S3) to house the migrated customer accounts. While both IoT platforms existed in parallel, OST was able to migrate thousands of devices to AWS with zero downtime.

“OST’s IoT Foundation was a pre-written, service-based platform with configurations for dozens of different domains and workflows at the API layer—that’s hard to come by,” Tarver said. “At the strategic level, the challenge was finding a partner like OST who could make the frameworks and AWS serverless components work together in concert. For us to do that work out of the box, it would take three years, instead of three months.” And finally, OST built the customer-facing website and application (for both iOS and Android) that interfaced with the IoT-connected grills.

Once re-platformed on AWS, Traeger was able to double the number of IoT-connected grills to 300K in just 18 months and can now scale to accommodate millions more. With currently 150 terabytes of data in storage, the company estimates its IoT data will reach 2 petabytes by 2021 and 3 zettabytes by 2023. Fortunately, the AWS IoT architecture put in place by OST will help Traeger tackle this classic big-data problem. “The advantage of AWS and a partner like OST is that these cloud and IoT services are pre-architected to handle and consider growth so we can scale,” Tarver said.

Traeger leadership initially set a target of $6 per device for the cost of replatforming its IoT grills. Several weeks into the migration, however, that target moved to $2 per device and then was reduced to $0.50 per device. By being strategic with a serverless architecture—leveraging Amazon API Gateway, AWS AppSync, and AWS Lambda—Traeger was able to optimize cloud spend and is looking into containerizing compute resources that handle persistent IoT requests for further savings. “We’re on track to reach our new target, and we’ve found substantial savings by continuing to rearchitect with AWS,” Tarver said.

Through machine learning to dig deeper into its IoT data, Traeger plans to develop predictive models that can signify part replacements and ultimately boost sales. Improved insights around grill usage by region and consumer will help the company better predict wood pellet sales, which is a major revenue source. Prescriptive modeling will enable Traeger to personalize its customer experience and recommend new product uses or recipes.