The Cloud Defined, Part 5 of 8: Measured Service

I am continuing my series of articles where I share my interpretation of the National Institute of Standards and Technology’s special publication 800-145. This week the focus is on the fifth and final fundamental characteristic of cloud computing according to the NIST’s definition:

Measured Service. Cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service.

The analogy that I use to describe this characteristic is McDonald’s. Yes, the golden arches fast food restaurant franchise is the perfect example of a measured service. A customer who enters a McDonalds sees a menu from which an order can be made. Items, prices, and some additional information are readily available to the consumer via that menu. You know what you are getting and precisely what it is going to cost.

But that is just the beginning of the process. McDonalds not only has a menu, but the company also knows exactly how much each item on that menu costs the organization, how much profit each item generates, what the process is for the production of each item, and so on, and so on. The secret of McDonalds is not the food itself so much as the company’s ability to consistently recreate the same food items (and customer experience) anywhere in the world. Every ingredient, portion, container, and other essential component of the McDonalds product line is measured and accounted for in precise detail for the company to recreate an item with.

This allows McDonalds to keep the most profitable menu items in supply according to consumer demand, and to also remove or revise less profitable items with exact knowledge as to how such decisions impact the bottom line. While I have focused primarily on the food in my analogy McDonalds does this for every component of their franchises. That is why every McDonalds feels familiar, because no matter which location you visit you are always receiving the product and customer experience that McDonalds has created for you by design.

This is what is happening with cloud infrastructures. Every component is measured to deliver a precisely configured product to the consumer. CPU performance, memory utilization, available applications, network bandwidth, and any aspect of your cloud solution is measured. This allows the IT staff to deliver a service that performs at a level that meets the end user’s expectations, while at the same time provides upper management with accurate numbers with which to measure the total cost of ownership and return on investment generated by the infrastructure.

Be sure to return next week as I cover the different types of service models available in a cloud computing infrastructure, and to continue my interpretation of the NIST’s definition of cloud computing.