Housing - Ein sinnvoller Weg für monolithische Anwendungen

Today I would like to take up the cudgels for housing. Housing means that servers are located centrally in a suitable data center, but are assigned to a dedicated user or application scenario.

Operating costs are crucial when choosing between hosting your own servers and using a public cloud platform. Public cloud platforms offer the convenience of scalable infrastructure and access to a variety of services, but they can often result in higher operating costs for non-scalable, monolithic applications with consistent hardware requirements. In these cases, housing your own servers can make more economic sense.

One of the primary benefits of housing your servers is the ability to control the hardware and software configurations. This makes a particular difference when discussing the operation of classic, monolithic applications such as ERP systems. These systems rarely are designed to be cloud-native and cannot scale flexibly across multiple instances. For the operation of such applications, a short-sighted lift & shift action to a cloud VM makes little sense and often drives up costs unnecessarily. It is often forgotten here that components such as virtual firewalls, virtual switches or traffic are billed in real-time on the familiar public cloud platforms, which can result in high ancillary costs. However, you should be honest and note that the space in a data center also costs money and that your servers consume electricity. Still, these costs have much lower variability than those of a cloud platform.

Another critical advantage of housing your own servers is the increased security and privacy. With control over the physical environment and access to the hardware, the risk of data breaches and unauthorized access is reduced. This can save on costly remediation expenses that may result from security incidents.

For non-scalable, monolithic applications with consistent hardware requirements, housing your own servers can make more economic sense regarding operating costs than using a public cloud platform. With the ability to control hardware and software configurations, negotiate better pricing, and reduce security and privacy risks, owning your servers can be a more cost-effective option in the long run.

My point here is not to pit the approaches against each other. Instead, these are two disciplines of the same sport, and our task is to select the right approach for the customer's specific requirements.