Unleashing the Power of Azure Kubernetes Service for Businesses

The best-known containerization platform is Docker, but it's not the only one on the market. Containerd is another competitor that has gained wide acceptance in recent years, even becoming the container platform used by the latest versions of Kubernetes.

Kubernetes

Now that we know what a container is, let's talk about Kubernetes.

Companies like TP require a flexible and scalable infrastructure to handle millions of customer requests per second. Kubernetes, a container orchestrator, plays a crucial role in monitoring, creating, deleting, balancing, and managing containers efficiently.

Docker is not the only containerization product, and Kubernetes is not the only available orchestrator. Other examples include OpenShift and Helios. According to a 2021 survey by CNCF, Kubernetes is used or evaluated in 96% of companies. Its high adoption is attributed to its rapid evolution and the support from its large development community.

Kubernetes consists of a Control Plane and Workers. The Control Plane oversees
orchestration, while Workers run applications and services.

About AKS

Large cloud providers offer container orchestration services using their own implementations or customizations of Kubernetes:

1. Azure: AKS (Azure Kubernetes Service)
2. Amazon: EKS (Elastic Kubernetes Service)
3. Google: GKE (Google Kubernetes Engine)

If you can work with container orchestrators in these three clouds, you will find that most of their concepts and commands are similar. AKS is CaaS. When an AKS cluster is deployed, it creates a Kubernetes service and a virtual machine scale set. The Kubernetes service is the orchestration service, while the virtual machine scale set provides the infrastructure for the control plane, worker nodes, and applications.

The Kubernetes service has a low cost, including a free tier. However, the real costs are in the compute (VMSS) which is created in a separate resource group. The cost depends on the number of nodes, their family, and the total hours of activity.

In a Kubernetes installation on-premise, the Control Plane is fully managed by the cloud provider, Azure. Our control over it is minimal.

Now, let's look at what makes the AKS so scalable and why it is one of the components that we use at TP to deliver the best services to our customers.

Pods

Pods are units where applications are executed. They have CPU and memory allocations and can be thought of as mini virtual machines for running applications. It's important to distinguish between pods and containers. While a pod is typically used for each container (application), it's not a limitation or restriction. Multiple containers can exist within the same pod.

As mentioned earlier, an application can have multiple instances. For example, there are three pods running container A and four pods running container B. It's important to note that pods running the same container don't have to be on the same Node. AKS will handle scaling or de-scaling the pods based on configurable parameters and demand. By default, the application starts with three pods, but as demand increases, AKS can create additional pods. Load balancers are used to distribute requests among all the pods automatically, without the need for administrator intervention.

AKS supports another type of scaling that enhances its robustness. In addition to automatic pod scaling, AKS also supports horizontal scaling. This feature enables efficient resource utilization, saving both time and costs in administration.

A short case study

Considering what was mentioned in the previous paragraph, let's look at this example.

A company in the sales sector uses AKS for its services. The AKS Cluster has 12 active nodes for daily operation. During promotions, the operation grows, and the 12 nodes become insufficient. AKS can automatically increase the number of nodes to ensure optimal service. When the load decreases, it returns to the 12 instances required for day-to-day operation, optimizing costs.

The benefits of AKS for businesses

1. Time and cost-efficiency. TP significantly reduces the time and effort required for managing Kubernetes clusters with AKS. This efficiency translates into cost savings and more resources in enhancing experiences for customers.
2. High availability. AKS’s robust architecture ensures high availability for critical customer service applications. By distributing workloads across multiple nodes and regions, uninterrupted service is guaranteed even during peak periods or unseen challenges.
3. Security first. TP places a premium on security, and AKS aligns seamlessly with this commitment. With Azure’s comprehensive security framework, our containerized applications are shielded from threats, ensuring customer data integrity and building trust.
4. Community and ecosystem. Leveraging AKS, TP taps into the vibrant Kubernetes ecosystem and community. This allows us to access a rich set of tools and resources, fostering innovation in integrated business services, such as customer service solutions, and keeping pace with industry trends.

AKS empowers industry leaders like TP to navigate the complexities of container, orchestration with ease. By embracing AKS, we have elevated our integrated business service capabilities, demonstrating the transformative potential of this cutting-edge technology.

Stay tuned for part 2. See you in the next blog!