What lessons can be learned about what we have learned from Prime Video microser

[43927555Participant]

Prime Video’s Prime Video team had followed my path of Serverless First, where the first attempt to build things is built using Step Functions and Lambda calls. In their blog, they say that it was easy to construct as well, which is exactly what they wanted to achieve. If you’re looking for ways to build something, creating an initial prototype within a couple of days or even weeks is an ideal method. After that, they attempted to scale the service to handle large volumes of traffic. They discovered that certain state changes within their step functions were often too frequent as well as some busy calls that were made between AWS lambda function and S3. The team was able to reuse the majority of their code through combining it into a single microservice that runs for a long time that can be horizontally scaled with ECS and is executed using a lambda. This is just one of the several microservices in Prime Video. Prime Video app. It’s a problem that the developers referred to it as changing a microservice into a monolith transformation, even though it’s actually a microservice stage of refactoring, which is exactly what I advise people to take in my presentations on Serverless first. I’m not a fan of „Serverless Only”, and I advised that if your needs want to sustain high volume of traffic with lower latency, and greater efficiency and higher efficiency, you need to re-implement your prototype in an autonomously running, continuous-running container that is part of a bigger serverless-based event-driven architecture that’s what they implemented. If you build it using a microservice from the beginning with, it’s likely to be slower (especially since you will have to make a lot of decision-making decisions regarding how to create and manage it) in addition to being more difficult to modify as you determine the thing you want to achieve. amazon.com/code

[Author]

Posts