The world moves faster than ever. And moving fast is one of the most important criteria of today's IT companies. We see companies that can adapt to the real world and the customer's needs with high velocity. The expected time to market of features has shrunk from years and months to weeks and less. In order to cope with this, companies not only need to introduce new technology or throw more money at their business problem, but also rethink and refactor the way they operate at the core of their IT.
What does move fast mean in this context? What aspect does this include? And which methods and technology support this?
Moving fast is all about quickly adapting to the needs of the market and customers. If a new feature is desired or looks promising, how long does it take to get from the initial idea to the feature being in the user's hands? If new infrastructure is required, how long does it take from that decision to the running hardware? And do not forget, if a certain piece of software is developed with all that velocity, is there automated quality control in place that ensures everything will work as expected and not break the existing functionality?
In software development, most of these questions lead to Continuous Delivery and automation. Software that is being developed needs to be built, tested, and shipped in an automated, fast, reliable, and reproducible way. A reliable, automated process not only leads to quicker turnarounds but ultimately higher quality. Automated quality control, such as software tests, are part of the process. In modern software development, Continuous Delivery, automation, and proper testing are some of the most important principles.
Traditionally, infrastructure was a big bottleneck in most companies. Smaller companies often struggled to provide new infrastructure with a limited budget. Bigger companies mostly fail to implement fast and productive processes. For big corporations, in most of the cases the issue is not the budget but the implementation of the processes. It is not untypical to wait days and weeks for new infrastructure, due to approvals and overly complex processes that technically could have been finished in a matter of minutes.
Therefore, application infrastructure and how it is designed is an important aspect. Chapter 5, Container and Cloud Environments with Java EE, will show you the topic of modern cloud environments. Actually, we will see that it's not so much about whether cloud service providers are being used. Fast and productive processes certainly can be implemented with on-premises hardware. Rather, it is more a question of whether processes are implemented properly, using well-suited technologies.
Modern infrastructure needs to be set up in a matter of minutes, in an automated, fast, reproducible, and reliable way. It should adapt to changing demands without great effort. To meet this criterion, infrastructure should be defined as code, either with procedural scripts or in declarative descriptors. We will see how infrastructure as code impacts software development workflows and which technologies support it.
These demands will impact the way teams operate. It is no longer sufficient for development teams to just develop and let the operational teams deal with running the software and facing potential issues on production. This practice always leads to tensions and finger pointing once critical errors occurred in production. Instead, the common goal should be to deliver software that fulfills a purpose. By defining the required infrastructure and configuration as code, development and operations teams will naturally move together. This DevOps movement, a compound of development and operations, aims toward accountability of the software team as a whole. Everybody involved is responsible for customers being able to use proper software. This is more an organizational challenge than a technical one.
On a technical aspect, Continuous Delivery, as well as the 12-factor and cloud nativebuzzwords attempted to meet these demands. The 12-factor and cloud native approaches describe how modern enterprise applications should be developed. They define requirements, not only for the development processes but also for the way applications are run. We will look into these approaches, modern cloud environments, and where Java EE supports us, later in this book.