On average we, as a modern digital human being, look at our smart phones more than 200 times a day. Apart from network connection issues in the few remote urban areas that still exist, we expect servers to deliver the data demanded by our apps on every moment we look. And we are not the only ones: modern digital human beings are scattered around the globe. And think about the big companies gathering our data to draw their digital mental pictures to ‘personalize’ the banners they show to us. And so far, we only considered mobile traffic.
Imagine what this means being a server: you have to deliver your data within a reasonable time 24/7. For each request. No matter what.
Of course, you’ve configured all your hardware redundant. But this talk is not about hardware, it’s about how your software reacts to failures.
Failure is not an option in services that need to be available 24/7. In a service integration environment in which there are a lot of API’s living who in turn call other API’s or backend services, there is a lot that can, and inevitably will go wrong. How do you minimalize the impact of failure?
In this session, we will show you how to use and configure Twitter’s Finagle client. We will show you running Java code examples demonstrating how you build a resilient HTTP client and simulate failure live on stage! Finagle client is written in Scala, and can be used by any JVM-based language.