Companies have different infrastructure needs and different possible ways to solve. The choice of infrastructure depends on various factors, in particular, the safety and costs that the chosen model produces. The two most used models by companies are cloud computing and on-premise.
When a group of servers is held and privately controlled it is the on-premise model. When the model predicts hiring of data center resources from a service provider it is cloud computing.
On-premise computing (often also called on-prem) requires the company to host private data centers for its management where resources are managed similarly to the cloud. Cloud computing, on the other hand, is offered by external service providers that offer resources based on the needs of individual customers, sharing the various computing resources among multiple individuals or companies and offering better security and downtime than on private alternatives.
On-Prem solutions offer the whole company the possibility of having total control of their infrastructure, while the cloud infrastructure does not guarantee total control but is very advantageous in terms of costs and scalability.
At the same time, the peculiarities of the two different infrastructures are very different. A public cloud offers downward and upward scalability, a tariff model based on the effective use of resources, a very low risk of downtime and the possibility of having integrated and automatic data backups, but the assistance that the user has a stable and effective internet connection because if weak or absent it becomes impossible to reach the data center. An on-premise system instead is single-tenancy, allow to have particularly personalized hardware (ad-hoc) and offers the possibility of having complete control of the resources and privacy on the user's data that held, but need for investments seen and periodicals and more commitment because the maintenance of any software requires higher costs. However, both solutions ensure data security through secure cryptographic systems.