Cloud-native development at local speed

The popularity of cloud-native applications, along with the pressure to build faster, has led to sweeping changes in the software engineering field—and to the rise of DevOps practices. However, deploying applications to the cloud has brought a host of concerns that slow down developers. This problem is highlighted during the write-deploy-test phase of the dev cycle, also known as the inner loop, when applications are deployed in an environment similar to production to test them in real-world conditions. Converting applications to a set of linked services, packaging them into containers, or instructing the target cluster to deploy the application (and its dependencies) are important considerations. Moreover, Kubernetes, the de facto orchestrator on which cloud applications run, brings its own concepts that need to be understood, while not being essential to the core functionality of applications. This need increases the development effort just to get an app up and running, not to mention the slowness of the process itself, as containers are built and then deployed. What can be done to improve the day-to-day experience of developers targeting Kubernetes clusters? What can make this inner loop faster, and bring the focus back on code? In this session, we'll look at the friction points that slow development early in a project, and then we'll see where things can be improved.