Understanding Right Sizing

A goal of most computer systems is high CPU utilization. If CPUs have low activity, processing resources are being wasted. But if a CPU is running at 100 percent, performance will suffer due increased process switching. Thus, the goal is a high level of utilization, but not maximum utilization.

As companies move new products onto the cloud, the company has no way of knowing what the level of utilization will be. If a company launches solutions with too few servers, the solution’s performance will suffer. If instead, the company provides more servers than are needed, the company will be paying for unused resources.

The virtual and easily scalable nature of the cloud makes it easy for companies to right-size their resource needs. With the company’s optimal CPU utilization come optimal related costs within the cloud’s pay-as-you-go environment. With the cloud environment, a customer may pay for an hour of processing by 50 servers—or the same amount for one server for 50 hours of processing. It’s easy to right-size, in other words.