Stripe Architects for Observability at Scale

Stripe resolved this challenge with a scalable monitoring and observability approach that uses Amazon Web Services (AWS). With approximately 3,000 engineers across 360 teams producing 500 million metrics every 10 seconds, Stripe had to look beyond its traditional observability stack and architected a solution on AWS for observability at a massive scale.

Stripe accomplished this through both architectural and cultural changes. First, Stripe adopted AWS services such as Amazon Managed Service for Prometheus—a service that monitors and provides alerts on containerized applications and infrastructure at scale—and Amazon Managed Grafana, which provides scalable and secure data visualization for operational metrics, logs, and traces. The serverless nature of these services provides for automatic scaling and efficient data processing. Stripe also created a sharded and tiered storage system, which distributes data across multiple databases to enhance scalability and reliability and separates frequently accessed data from less accessed data to optimize storage costs.

Culturally, Stripe fostered a mindset of self-reliance and continual improvement among its engineers, emphasizing adaptability and efficient observability practices. The company also created user-friendly observability practices so that engineers could quickly get used to the new tools at their disposal.

“You need to really cultivate that culture of self-reliance and finally make it simple to do the right thing for your users,” says Cody Rioux, staff software engineer at Stripe.

Stripe’s new observability approach brought a multitude of benefits. With scalable AWS services, the company can efficiently adapt to increasing data demands without inflating costs. This approach has also improved visibility into Stripe’s applications, helping the company maintain consistent system performance. The managed services offered by AWS reduce the complexity and effort required to manage infrastructure, empowering engineers to focus on more important and innovative work. On AWS, Stripe can continue to evolve and scale its services without worrying about outgrowing its observability stack, which helps drive customer satisfaction and business growth.