Hybrid Cloud Strategy: Application Agility and Infrastructure Freedom with Kubernetes

hybridcloudkubernetescontainers

October 25

The rise of containerized applications has had one of the most powerful impacts on the infrastructure stack since the advent of virtualization. Kubernetes has introduced powerful application agility that the Evoque team celebrates—our Multi-Generational Infrastructure offering was created to accelerate innovation and give our customers flexibility to move their containerized applications seamlessly between the data center and the cloud.

How did open source Kubernetes (aka K8s) become the 800 pound gorilla of infrastructure architecture? A strong ecosystem of supporting tools and dominant base of engineers having achieved (and continuing to develop) fluency, has been key.. With the embrace of AWS, Google Cloud Platform, Azure, bare metal enthusiasts, and the VMWare and Red Hat crowd, it’s clear that containers are here to stay.  

Let’s wade into this revolutionary application design framework to learn more about containers and see why abstracting the infrastructure from the application can be a huge advantage in the cloud, in the data center or increasingly both.

kubernetes for hybrid cloud

Container Use in Hybrid Designs

In hybrid cloud technologies, containers are a widely used option for deploying and managing applications and software in the cloud. Containers extract applications from the physical environment in which they are running, containers manage all dependencies required to an application and execute them in an isolated environment. Containers follow the basic concept of virtualization, in which a virtual machine is created by creating a software layer between the operating system and the CPU. 

Virtual machines (VM) utilize physical drawbacks more efficiently, however, they have some limitations. Each VM creates a complete copy itself, which makes them more resource-intensive and problematic for applications that don’t require such functionality. Containers solve this problem; they are like virtual machines but they run like a subset of an operating system (OS) and consume just enough OS resources to run the application that they contain. 

After analyzing the features of Kubernetes, some vendors have decided to take K8s as the first approach to hybrid cloud. For example: 

  • Google Anthos utilizes Google Kubernetes Engine to maintain clusters running in hybrid clouds or private data centers; 
  • VMware’s Tanzu is managing their machines deployed on the cloud with the help of Kubernetes; 
  • Amazon’s Web Services (AWS) uses Amazon’s Elastic Kubernetes Service (EKS) to manage private and public clouds. 

K8s provide some essential features that are promising in the hybrid cloud computing environment. Containers can be killed, replaced, and self-healed automatically. The most robust feature of K8s is their scalability that can scale efficiently. They provide horizontal scaling of the deployed application, where it can be scaled as needed from command-line to user-interface. With automated rollbacks and rollouts, the health of deployed applications are checked regularly ensuring that all instances don’t fail or go down concurrently. If something bad happens, K8s rollback immediately. K8s have load balancing features that provide different IP’s to the containers behind a single DNS name to manage the load on the containers. The platform automatically troubleshoots problems as restart failed containers, replacing the node in case of a node failure, and hold to publish containers to the users until they are ready. 

Open-source Kubernetes will continue to improve and become even more performant and resilient. As an architecture to deploy applications, containers provide a set of standards to deploy and run an application in a manageable way no matter the infrastructure. While hybrid cloud deployments are not a primary focus of the Kubernetes, it has delivered huge advantages to the hybrid cloud practitioners by providing a systematic way to deploy and manage applications running in both the cloud and data center. To take a deeper dive into containerized applications and the benefits of Kubernetes, please download Evoque’s cloud engineering arm Foghorn’s whitepaper on the subject.