App modernization

Modernizing application development with serverless and Quarkus

William Markito Oliveira
© Shutterstock / Vichy Deal

It is no secret that our world is changing. This article will explore how using a combination of Knative, the Kubernetes-native serverless platform, and Quarkus, Red Hat’s container-native approach to Java, can be used together to help simplify the modernization of application delivery.

It may finally be safe to say that serverless is moving beyond being simply a buzzword and is becoming a key part of overall business strategy. It is rising in both popularity and business use cases, with companies increasingly recognizing the importance of having a solid serverless plan in place as part of their overall cloud strategy. However, despite the increase in use, serverless architectures and technologies are still seen as requiring a big shift in the way applications are built and deployed, and are often still tied to digital transformation projects. Many companies may not realize that there continues to be a big opportunity to be able to modernize and migrate legacy applications by running them in serverless containers, with no major change in the current development and deployment environment. This article will explore how using a combination of Knative, the Kubernetes-native serverless platform, and Quarkus, Red Hat’s container-native approach to Java, can be used together to help simplify the modernization of application delivery.

SEE ALSO: Quarkus – an IO thread and a worker thread walk into a bar: a microbenchmark story

The importance of app modernization

It is no secret that our world is changing. With the ongoing shift to digital-first, accelerated by the COVID-19 pandemic, it is clear that companies’ backend technologies need to keep pace. All customer transactions are being done digitally, across every industry. From both a competitive and customer experience perspective, things have to be running seamlessly. Modern, cloud-native and container-first technologies, such as serverless and Quarkus can help deliver software faster and without interruption. It’s true that app modernization is not a new IT priority, but it is more widely understood now that there is no wiggle room in performance. Companies may also be focused on AI/ML and automation, and having a hybrid cloud approach can make it0 easier to adopt these capabilities into their organizations. By leveraging the tools already in place, there is no need to hire new or additional tech talent.

Successfully adopting Kubernetes and container-first serverless

Despite only being five years old, Kubernetes has seen fairly significant adoption in both developer communities and in the enterprise, as a trusted, secure resource. Having a Kubernetes strategy in place, as developers know, opens the door up for a more robust container-first approach, as well as infrastructure-as-code, more innovation with cloud-native tools, flexibility, portability and access to community tools, among other aspects. While Kubernetes rose to popularity because of its ability to work with containers and manage virtualization, abstraction and complexity, its future may very well depend on how well it is able to adapt to serverless, which is where Knative serving and eventing comes in.

Where it once may have been viewed as a luxury, having a proper technology strategy in place is now seen as a necessary utility, much like electricity is to a house. A serverless approach helps abstract complexity when delivering on the “technology-as-a-utility” premise, and also helps modernize application development and deployment. If your team does not have a serverless strategy in place, you are most likely spending quite a bit of time in architecture and design meetings, discussing capacity planning and using traditional application servers to build applications. This is a tedious and time-consuming process, where developers are spending hours in meetings instead of building applications. Having a serverless and cloud-native approach in place will both empower developers to be able to speed up Kubernetes adoption, quickly spin up new applications, and further scale. When using a successful community project like Knative as the foundation, you can have more confidence adding serverless to an existing Kubernetes platform, without lock-in concerns.

SEE ALSO: Micronaut benchmarks: Faster than Quarkus and Spring Boot?

How does Quarkus fit with serverless?

For many developers, Java is the foundation of their work, and application modernization can’t mean leaving Java behind. The Red Hat Build of Quarkus is our Kubernetes-native Java framework, that allows Java developers to continue working in the language they know and love, while being able to work in modern, cloud-native environments, such as serverless. Historically, Java has been too slow at startup and too memory consuming to be able to be used in serverless environments. Quarkus makes Java ready for containers, by reducing the memory needed at startup, offering near-instant scale-up times, and achieving a much higher workload density. It also helps boost developer productivity through live reload features and is compatible with popular frameworks like Eclipse MicroProfile and Spring. Bringing Java into the future of application development is a key factor in overall app dev modernization, and Quarkus supports this goal.

By having the right tools in place, the process of modernizing applications to better fit into our new, all-digital business landscape can be simplified. Serverless architectures mean a streamlined, modern application development process, and by allowing Java developers to use the language they already know and love, with no additional training necessary.


William Markito Oliveira

William Markito Oliveira is an energetic and passionate Technical Product Manager with expertise in software engineering and distributed systems. He leads a group of product managers working on innovative and emerging technologies, from inception to market, acting as a proxy for customers, business units, and engineering teams.

Inline Feedbacks
View all comments