Blogs›Stripe›Microservice Testing with Apache Spark
This feature thread tracks Stripe's advancements in microservice testing, particularly focusing on leveraging Apache Spark to address the challenges of testing services with complex input dependencies. The initial posts introduced the concept of using Apache Spark to build repeatable replay harnesses for testing microservices that are difficult to model by hand. This approach aims to improve test coverage and efficiency by allowing for the replay of a long tail of inputs without requiring manual modeling. This post details how to scale up this approach using Spark's parallelism and linear scalability to build regression test harnesses, enabling comparison of old and new implementations, modeling rule changes, and quantifying impact before production.
6/8/2026
This post continues the discussion on using Apache Spark for microservice testing, likely detailing further advancements or optimizations in building repeatable replay harnesses. It builds upon the previous post's introduction to the concept, potentially offering more in-depth technical details on implementation, performance tuning, or strategies for handling a wider range of complex input dependencies.
6/1/2026
This post details how to scale up microservice testing using Apache Spark's massive parallelism and linear scalability. It explains how to leverage Spark to build regression test harnesses that can compare old and new implementations, model upcoming rule changes, and quantify impact before changes reach production, addressing the challenge of testing microservices with a long tail of difficult-to-model inputs.