Not all cloud computing is the same. It depends on the right operating model.

If you talk to software designers and IT managers about current industry trends these days, terms like artificial intelligence, extended reality, hyperautomation, distributed cloud, multi-experience, 5G and many more quickly come up. Without a doubt all exciting developments with great future potential.

But if you look at what is already being implemented here and now, one issue in particular comes into view: cloud computing. The shift of IT resources away from local workgroup servers and corporate data centres to remote data centres of cloud providers is a mega trend of our time. It’s happening now.

Drivers of cloud computing

The motives for this are easy to understand and follow familiar patterns: Reduction of capital and personnel costs, demand-oriented provision of required resources, increased flexibility, softening of rigid limitations to use in terms of location, time and available end devices, etc.

Nevertheless, it is important to look closely, because a number of different operating models have also become established in cloud computing. And not every model fits every need.

There are three operating models in the foreground: Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS).

To understand them, it is worth taking a look at the classic operation of server applications at companies (on-premise). The following figure shows the typical software and hardware stack, i.e., the various components that mesh together.

Structure of IT systems

At the top is one or more mostly industry-specific applications and the data they manage and process. This is the level with which the end user interacts.

Below that is the middleware, software that performs database functions, transaction processing or, for example, authentication and encryption for various applications.

Below that is the operating system as an intermediary between the software and the hardware, often followed by a virtualisation layer. It has the task of running several different operating systems or operating system instances in parallel on one physical piece of hardware. This allows for better utilisation of the available hardware resources.

Only then does the actual server hardware follow: CPU, memory, data storage and network connections.

The various cloud models differ in terms of which of these layers are managed by the cloud provider and which are under the control and supervision of the customer wanting to move their IT resources to the cloud.

• With classic user-managed server operation (on-premise), the operator/company is responsible for all these levels. If additional memory is needed, if a hard disk fails, if the operating system or other components need to be updated – all this is a job for the respective IT department.
• With Infrastructure as a Service (IaaS), the cloud provider is responsible for the provision and maintenance of the hardware and for the virtualisation. The customer can install any operating systems, middleware and applications on the platform provided remotely via the internet and is also responsible for their operation and maintenance. They do not come into contact with the hardware.
• Platform as a Service (PaaS) goes one step further. Here, the cloud provider also installs and configures the operating system and middleware for its customers, monitors their operation and regularly installs all necessary updates and releases. On this clearly defined basis, the customer installs its applications and controls their operation. The customer doesn’t have to worry about anything else.
• Software as a Service (SaaS) is one level higher still: SaaS provides a complete product that is run and managed by a service provider. In most cases, SaaS refers to end-user applications.

With SaaS, customers no longer have to worry about setting up and maintaining the service. It also does not matter which software and hardware components are required for this or how the underlying infrastructure is managed. They can concentrate solely on using the respective software. This is therefore often the preferred model with industry solutions, and this suit fits perfectly.