IT professionals hear a lot about right-sizing systems on cloud platforms, but many companies are still choosing to utilize on-premises hardware to set up massive big data systems and other enterprise architectures, which they will right-size in order to promote efficiencies.
One of the biggest reasons for choosing on-premises systems is that a large and established businesses may have the upfront capital to put into hardware acquisition. In this type of scenario, the company may actually feel they are better positioned by making that upfront capital outlay rather than paying cloud service fees on an ongoing basis.
In addition, there is also a great sense of control and additional security for companies that are maintaining systems on-premises. There’s been a lot said about the dicey security of some public cloud operations. Companies may be wary of handing over their data to a vendor, and unsure about how a vendor will keep data safe. Other nightmare scenarios involve vendor lock-in and cases where vendors can hold data hostage to prices or SLA contractual obligations.
Companies can utilize various costing tools to look at the overall costs of either cloud or on-premises system development. They can also use popular brand name services like Amazon Web Services and Microsoft Azure to run with on-premises systems. For example, Amazon Elastic Compute Cloud or EC2 is designed to run with AWS cloud platform systems, but can also be used to run instances in an on-premises data center. Some companies get a real feeling of safety and security from utilizing these brand-name vendor options in an on-premises environment.
As for right-sizing, companies will have many of the same goals that they would have right-sizing in the cloud. In fact, in an on-premises environment, the desire to control resource use is much more urgent. Where cloud-based systems can be scaled up simply by utilizing on-demand services, scaling in an on-premises environment requires much more in terms of logistics. That said, the above reasons will compel many companies to do the work required to manage right-sizing in an on-premises data center rather than simply outsourcing much of the data handling that supports business to a cloud vendor.