Understanding what cloud native architecture is

Different organizations have different definitions of cloud native architecture. Almost all definitions emphasize creating scalable, resilient, and maintainable systems. Before we proceed to a cloud native architecture definition, we need to understand what cloud computing is about.

Cloud computing

The simplest introduction to the cloud is the on-demand delivery of infrastructure, storage, databases, and all kinds of applications through a network. Simply, the client outsources computation to a remote machine instead of doing it on a personal computer. For example, you can use Google Drive to store your files or share images with Twitter or Firebase applications to manage mobile application data. Different vendors offer services at different levels of abstraction. The companies providing these services are considered to be cloud providers. Cloud computing pricing depends mainly on the utilization of resources.

The cloud can be divided into three groups, depending on the deployment architecture:

• Public cloud: In the public cloud, the whole computing system is kept on the cloud provider's premises and is accessible via the internet to many organizations. Small organizations that need to save money on maintenance expenses can use this type of cloud service. The key issue with this type of cloud service is security.
• Private cloud: Compared to the public cloud, the private cloud commits private resources to a single enterprise. It offers a more controlled atmosphere with better security features. Usually, this type of cloud is costly and difficult to manage.
• Hybrid cloud: Hybrid clouds combine both types of cloud implementation to offer a far more cost-effective and stable cloud platform. For example, public clouds may be used to communicate with customers, while customer data is secured on a private network. This type of cloud platform provides considerable security, as well as being cheaper than a private cloud.

Cloud providers offer services to the end user in several ways. These services can rely on various levels of abstraction. This can be thought of as a pyramid, where the top layers have more specialized services than the ones below. The topmost services are more business-oriented, while the bottom services contain programming logic rather than business logic. Selecting the most suitable service architecture is a trade-off between the cost of developing and implementing the system versus the efficiency and capabilities of the system. The types of cloud architecture are as follows:

• Software as a Service (SaaS)
• Platform as a Service (PaaS)
• Infrastructure as a Service (IaaS)

We can generalize this to X as a service, where X is the abstraction level. Today, various companies have come up with a range of services that provide services over the internet. These services include Function as a Service (FaaS), Integration Platform as a Service (iPaaS), and Database as a Service (DBaaS). See the following diagram, which visualizes the different types of cloud architecture in a pyramidal layered architecture:

Figure 1.5 – Types of cloud platforms and popular vendors

Each layer in the cloud platform has different levels of abstractions. IaaS platforms provide OS-level abstraction that allows developers to create virtual machines and host a program. In PaaS platforms, developers are only concerned about building applications rather than infrastructure implementations. SaaS is a final product that the end user can directly work with. Check the following diagram, which shows the different abstraction levels of cloud platforms:

Figure 1.6 – Levels of abstraction provided by different types of platform

SaaS offers utilized platforms for the developer to work with where developers just need to take care of business logic. The majority of the SaaS platform is designed to run a web browser, along with a UI to work with. SaaS systems manage all programming logic and infrastructure-related maintenance, while end users just need to concentrate on the logic that needs to be applied.

In SaaS applications, users do not need to worry about installing, managing, or upgrading applications. Instead, they may use existing cloud resources to implement the business logic on the cloud. This also decreases expenses, as the number of individuals who need to operate the system is decreased. The properties of SaaS applications are better suited to an enterprise that is just starting out, where the system is small and manageable. Once it scales up, they need to decide whether to hold it in SaaS or switch to another cloud architecture.

On the PaaS cloud platform, vendors have an abstract environment where developers can run programs without worrying about the underlying infrastructures. The allocation of resources may be fixed or on demand. In this type of system, developers will concentrate on programming the business logic rather than the OS, software upgrades, infrastructure storage, and so on.

Here is a list of the advantages of building cloud applications using PaaS:

• Cost-effective development for organizations since they only need to focus on business logic.
• Reduces the number of lines of code that are additionally required to configure underlying infrastructure rather than business use cases.
• Maintenance is easy due to third-party support.
• Deployment is easy since the infrastructure is already managed by the platform.

IaaS is a type of cloud platform that provides the underlying infrastructure, such as VMs, storage space, and networking services, that is required to deploy applications. Users can pick the resource according to the application requirements and deploy the system on it. This is helpful as developers can determine what resources the application can have and allocate more resources dynamically as needed.

The cost of the deployment is primarily dependent on the number of resources allocated. The benefit of IaaS is that developers have complete control of the system, as the infrastructure is under the control of the system.

A list of the advantages of using IaaS is given here:

• It provides flexibility in selecting the most suitable infrastructure that the system supports.
• It provides the ability to automate the deployment of storage, networking, and processing power.
• It can be scaled up by adding more resources to the system.
• The developer has the authority to control the system at the infrastructure level.

Although these fundamental cloud architectures are present, there are additional architectures that will address some of the challenges of cloud architectures. FaaS is once such architecture. FaaS operates pretty much the same as the PaaS platform. On the FaaS platform, the programmer provides a function that needs to be evaluated and returns the result to the client. Developers do not need to worry about the underlying infrastructure or the OS.