Matt Stine

Cloud-Native Application Architecture

Monday, 9:00 AM EST

Cloud-native architectures combine the unique aspects of cloud platforms with the principles of DevOps and Continuous Delivery to enable the rapid development, deployment, and management of applications. As the speed of innovation becomes one of the key drivers of business success, these architectures ensure teams are able to meet the need of the business, to move quickly, while at the same time ensuring important non-functional characteristics like availability and scalability.

Many of the innovators in this space, including Amazon, Twitter, LinkedIn, and Netflix, leverage small, autonomous teams which focus on business capabilities and build twelve-factor style, microservice applications. Microservices integration is achieved via lightweight, decentralized, and choreographed point-to-point interactions rather than the heavyweight, centralized, and orchestrated ESB-style integration found in traditional SOA.

With the advent of cloud-native architectures, building distributed systems will become increasingly common for the enterprise Java developer. Fortunately many of these same innovators have embraced the JVM as they’ve built increasingly complex systems, with Netflix open-sourcing much of its toolkit for constructing these systems at NetflixOSS.

Cloud Foundry and Spring provide open source framework tooling and platform services for developers to quickly build some of the common patterns in found in distributed, cloud-native systems. Many of these patterns are provided by the Spring Cloud project, which wraps many of the battle-tested components found at NetflixOSS with the Spring programming model, and provides easy deployment of NetflixOSS services to Cloud Foundry.

In this class the learner will have the opportunity to practice working with cloud-native architectures using Spring and Cloud Foundry.

Refactoring the Monolith: LIVE!

Wednesday, 9:00 AM EST

Many of us would love to embrace microservices in our day-to-day work. But most of us don’t have the opportunity to start over with a pure greenfield effort. We have to understand how to refactor our existing monolithic applications toward microservices. Practical steps include building new features as microservices, leveraging anti-corruption layers, strangling the monolith.

In this presentation we’ll go light on the theory and walk through the actual process of turning a strawman monolith into a family of well-factored microservices.

Microservices in the Large: Tracer Bullet Architecture

Wednesday, 11:00 AM EST

Much is said about the decentralized governance of and local autonomy given to “two pizza teams” build microservices. But how do you organize teams to effectively collaborate to build the eventual composite system?

In this presentation we’ll examine how to apply the Tracer Bullet Development methodology described in Ship It! to effectively construct distributed systems composed of microservices.

Microservices in the Small: Test-Driven Design

Wednesday, 1:30 PM EST

While microservices may be smaller than some of their architectural cousins, that doesn’t mean that testing the internal components of a microservice takes a back seat.

In this presentation we’ll walk through the test-driven workflows that you can use to build a well-factored microservice.

Distributed Tracing of Microservice Architectures

Wednesday, 3:15 PM EST

Embracing microservices also means embracing distributed systems. Distributed systems carry with them multiple challenges. One set of challenges includes problem of visibility into the behavior of the composite system, understanding that behavior, and being able to isolate the cause(s) of problematic behavior. These challenges can be addressed by applying the techniques known collectively as Distributed Tracing.

In this presentation, we’ll examine the theory of distributed tracing put forth in Google’s Dapper paper, and we’ll look at how this theory is put into practice in the design of Zipkin, an OSS distributed tracing platform.