Orange Theory Fitness – Boca Raton FL

Orangetheory is a 1 hour, full body workout, focused on training Endurance, Strength and/or Power.  We use Heart Rate Based Interval Training, which burns more calories post workout than a traditional exercise. When wearing our Heart Rate monitors, your real time results are displayed on large screens throughout the studio. Intensity is based on your individual Heart Rate zones, making the workout effective for all fitness levels. To top it off, our fitness coaches to lead the workout to prevent you from over or under training.

 

Position

At OrangeTheory Fitness I served as the solutions architect. Charged
with the complete redevelopment of the existing ad hoc infrastructure
to a scalable cloud based solution.

 

OTF was challenged with a large amount of active users (over half a million).
Yet their function comes from the ability to sync all possible values for a heart rate reading
during a session.

 

Challenges

  • Development team was based in Vietnam
  • Executives desired moving to AWS
  • Massive amounts of legacy data, stored in disparate sources.
  • Lack of any central membership authority.
  • Ingesting large amounts of sensor data from heart rate motors.

 

Solutions

 

DynamoDB quickly rose to be a top contender for storing our ever growing properties of member data. Making it relatively easy and inexpensive

 

Lambda Event driven processing allowed OTF to begin the process of both centralizing and streamlining the ability to introduce revolutionary function

 

S3 Continual archive of all sensor and gym equipment data ever created. This data could be loaded into virtual Redshift instances for intense meta analysis.

 

SQS Distributed machine control. Such as pausing or starting a class within a gym room.

 

SNS Event based processing allowed OTF to spare the cost of creating a server farm and worrying about redundancy. Leaving us the ability to focus on the solutions for problems unique to OTF that justify the payroll.

 

MindBody Online
Implementing both interval and event driven data synchronization. This was utilized specifically to address the ad hoc membership problem. By creating our own member store in DynamoDB and updating it on data change through exposing a Javascript Lambda API through the AWS Gateway. As a solution architect I was responsible for signing off on final technical requirement documents between OTF and MBO.

 

Projected Future
All machine and sensor data will flow seamlessly on an as needed basis. OTF will be able to preserve valuable metadata that would otherwise be too unwieldy to store for any significant amount of time, never mind attempt to read and analyze for functional and corporate benefit.

Complete synchronization between the independently managed gyms and the parent OTF. A single membership repository with an adapter layer that allows the development teams to easily introduce and maintain 3rd party source updates.

Dynamic Frontend that renders in both web, desktop, mobile, and on tv screens within locations.

 

 

One can conclude that AWS was a smart choice for OTF at the time. I always took advantage of our creative ability to wrap AWS in such a way that we can pull out our assets (code, etc) and deploy on another platform in the future.