Roy Rapoport explains the concept of canary analysis and how Netflix uses it to deploy software to its internal systems.
Ashley Puls explains Java bytecode manipulation: reasons for manipulating bytecode, libraries that help, how NewRelic is using it, and more.
The OSGi Alliance is working on a Promises specification which will provide CompletableFutures that can run all the way back to Java 1.4. Tim explains why this is of benefit and how it compares to Java 8, along with where you can find out more information about the project and how it can be used both inside and outside an OSGi framework.
Engine Yard's Ines Sombra discusses state management in the cloud in general, and specific data stores including MySQL, Postgres and some NoSQL alternatives. She also explains why the pets and cattle analogy doesn't work for her, and what need to be done in organisations with respect to trusting people and trusting the infrastructure.
Craig Motlin, technical lead of the GS Collections project, talks about where GS Collections came from, how it compares with other collections libraries, and what influence it had on Java 8. He describes the different philosophy of GS Collections as compared to other collections libraries, and what benefits open-sourcing the internal library has had
Ian Robinson talks to Charles Humble about the history of Neo4J, it's data structure, and use cases such as recommendation engines, network impact analysis, route finding and fraud detection.
Tim Ward speaks to InfoQ about Bndtools, a means of building OSGi solutions in Eclipse using a code-first approach. Bndtools provides a means to build, test, automatically version and deploy bundles. By providing a bundle-on-save action, Bndtools can automatically create the build JAR and (re)deploy it into a running framework, giving one of the shortest development cycles available in IDEs today.
Adam Ernst talks about how functional programming and immutable data structures have made Facebook’s iOS app much easier to test and debug. By decoupling the data pipeline from the UI objects, and minimising the wrk on the UI thread, the application has become easier to test and suffers less bugs than when the UI was generated procedurally.
Yoni Goldberg explains Gilt's architecture which consists of 350+ microservices, how teams decide the scope of a microservice, API design and management, monitoring, Scala at Gilt and much more.
Dave McCrory explains how he coined the term 'data gravity', and how he expects the rise of data related microservices to deal with its consequences. He also gives an overview of Basho's Riak version 2, and what else can be expected from that platform in nearby releases.