To create amazing cloud native applications your platform needs these components: Kubernetes, Continuous Delivery Pipelines, Edge Stack, and Observability. This article will explain the importance of each component.
Building Cloud Native Applications
Companies are increasingly looking to digital transformation to enable their teams to iterate on and deliver new functionality to their customers faster. For many organizations, supporting these changes will require the adoption of modern infrastructure and tooling, as well as a cultural shift.
This article focuses on the need to implement four key elements of a modern infrastructure platform, which we have identified from working with 1000s of users that have adopted the cloud native approach
To build cloud native applications, your developers need access to a cloud native platform that meets their requirements for shipping features with speed and safety.
This platform must enable full cycle development so that your developers can ship code in a self-service manner. To achieve this, all cloud native platforms must provide four key components:
- Kubernetes
- Continuous Delivery Pipelines
- Edge Stack
- Observability
1. Kubernetes
Containers have become the standard unit of cloud deployment for three main reasons:
- Containers require less resources to run than VMs
- The Dockerfile manifest format enables developers to define their own build and deploy configuration
- The “docker build .” method makes it easy for developers to assemble applications in a self-service manner
However, with containers come more operational concerns, like restarting and rescheduling containers when underlying hardware fails. To address these concerns many teams turn to a container orchestration framework like Kubernetes.
Kubernetes was originally released by Google in 2014 and has been widely adopted as a container manager since then. Used in production by many organizations, Kubernetes provides a strong foundation to support the other three capabilities of a cloud native platform. Kubernetes gives both developers and platform teams a standardized workflow and toolset.
2. Continuous Delivery Pipelines in Cloud Native Applications
Continuous delivery pipelines enable developers to deliver new features, bug fixes, and experiments into production as rapidly and safely as possible. With a continuous delivery pipeline, a developer is able to push an application deployment to production on-demand without incurring the operational overhead that may exist with a monolithic application.
Progressive delivery is an extension of continuous delivery that uses cloud native traffic control mechanisms and advanced observability tooling to enable developers to ship features incrementally and measure their impact with real user-traffic. The successful implementation of both continuous delivery and progressive delivery is dependent on developers being able to define, modify, and maintain pipelines that codify the applications’ build, quality, and security assertions.
3. Edge Stack
The Kubernetes cluster edge is a critical component of any Kubernetes or cloud native application. An edge stack is the set of proxies that sit between our application and your end users. Commonly, a cloud native edge stack is made up of an ingress controller or an API gateway, a content delivery network (CDN), a layer 7 load balancer, a web application firewall (WAF), and an identity aware proxy.
Cross-functional requirements are controlled at the edge including:
- Security - authentication, transport level security (TLS), and denial of service attacks
- Reliability - rate limiting, circuit breaking, timeouts.
Additionally, it is critical that your cloud native edge stack integrates with your progressive delivery pipelines in order to enable the controlled release of applications. Many external tools (ex. Identity providers or continuous delivery pipelines) are not useful until they are configured in a way in which your application is exposed to the internet via your ingress controller.
In order for developers to be able to configure the edge in a self-service manner, the edge stack must offer comprehensive support for a range of protocols, architectural styles, and interaction models that are commonly found with a polyglot language stack.
4. Observability in your Cloud Native Apps
With more complex systems comes a greater need for visibility into the systems themself. Implementing observability for cloud native applications is inherently more challenging than for monolithic applications. Leaders in the space have defined the three pillars of cloud native observability as monitoring, logging, and distributed tracing. An effective observability stack addresses all three of these pillars holistically.
How to Configure Your Own Cloud Native Platform
Many teams will spend months or years iterating on the different elements of their cloud native platform looking for the best solutions. However, if you’re new to Kubernetes it is important to deploy and configure an application-ready Kubernetes cluster as quickly as possible so your engineering teams can experiment with technologies, workflows, and exposing your application to real user-generated traffic.
The K8s Initializer provides a web interface for developers to get an application-ready Kubernetes cluster up and running in just a few clicks.
With options to configure all of the four key components of a cloud native platform, the K8s Initializer takes care of customizing your configuration to your environment -- a process that typically involves lots of searching on StackOverflow and then combing through hundreds of lines of YAML by hand.
Be sure to check out this tool when building your own cloud native applications!