Capacity and the Cloud

Have you ever tried to place an order online only to see that the “server is too busy”? This is a frustrating example of where a service provider is having capacity issues that are negatively impacting their ability to do business. So what is capacity management and how can cloud technologies help?

Before we dive into capacity, let’s first look at demand. Demand is the need for resources to perform some piece of work in a given period of time. For example, it will take five people to dig that hole in one day. Capacity is the ability to meet the demand with the appropriate resources. I have five people therefore the hole will be dug in one day.

Unfortunately, most organizations face resource constraints and struggle to provide the exact amount of resources at the right time to meet demand. They often rely on some form of guessing to determine demand (forecasts and models) and then secure resources based on these guesses. If you manage a restaurant and are having a slow, you can send staff home and shed extra resources that are going underutilized. However, if you are in technology and forecast business transactions and then purchase infrastructure (capacity) to handle the projected business (demand), you can’t just send servers home early on a slow night. Now you are in a position where you are over capacity and that can be expensive. On the flip side, if you get a sudden unpredicted burst in business, your users may get the “server is too busy” message if you are under capacity and you lose business. Getting this balance right can be quite challenging.

What if you had a flexible resource pool that you can change as business conditions change so that you can always have exactly what you need when you need it? In the restaurant, you have a team of people sitting idly by (at no cost to you) and you can ask them to come in on a busy night or send them home on a slow night.

That is exactly the model today’s cloud service providers are making available to businesses now. Elastic infrastructure that can adjust capacity resources quickly based on actual demand monitoring. When done properly, you can actually have your infrastructure grow and shrink just ahead of demand and always have exactly the right amount of capacity to meet demand. Who wouldn’t want to take away the risk and expense and do it this way?