Figure 1. On-premises computing characteristics
The following are the basic components of an on-premises (that is, a traditional Data Center) compute environment:
- Equipment is procured using a Capital Expenditure (CAPEX) model. This means that purchases are planned, and monies budgeted and set aside for the project (typically aligned with the reporting periods of the organization).
- The equipment is housed in a dedicated facility on the customer’s premises. This means the customer is responsible for the physical plant, the power supply, the HVAC system, physical perimeter security, some sort of fire detection and suppression system, and so forth.
- A team of employees or contractors is used to perform system administration, management, planning, and user support services. Notably, this team responds to user requests for assistance and maintains a trouble ticketing system for administering the user support experience.
- The IT department responsible for this infrastructure provides ongoing maintenance, notably the hardware and software upgrade and refresh cycles, software maintenance and updating, repairs and replacements, change management, and QA testing.
- Services are provided to users in several forms: as an application service (such as a web portal, email server, or LOB application), as bare metal servers, as virtual machines, or as containers.
- The IT department manages their own internal network for interconnection of the equipment in the Data Center and to users throughout the organization, including a network connection to an Internet Service Provider (ISP) or potentially to a dedicated network connection to a Telecom provider.