The cloud is an ideal business solution for organizations that have unpredictable demand. To understand why this is, we need to briefly discuss the alternative case, in which the cloud is not used and a business must provide its own servers. In such a case, the business must do one of two things: either have enough servers and computing power available to serve peak demand even at times of low demand, or else at times of peak demand turn away some demand. Although either option can work, neither option can work in the situation here. The former option can work where the times of high demand are frequent enough to pay for the costs associated with having some idle time of low demand.
This is because cloud computing is, per server-hour, much more expensive than owning your own machine and running your website on that machine. The cloud provider has a healthy profit margin that they add to the cost of hiring cloud computing from them. Of course, this may not always be true (for instance, if you pay significantly more for electricity than the cloud computing provider does), but in general, the principle holds. In some cases (for instance, with Microsoft Azure), the convenience achieved through the use of cloud computing allows the provider to charge enormously more than it would cost to achieve similar results yourself. So, all in all, if demand is unpredictable but usually very high, it may be beneficial to have your own server.
Use your promo and get a custom paper on
"Cloud Computing Business Case".
Similarly, if you are in a business where demand is unpredictable but usually low, and has very infrequent peaks, it may be acceptable to simply return an error when people attempt to access the site during its peaks. An example of such a situation is a website that provides information that is useful to people within a small niche, and valueless but possibly interesting to everyone else. For instance, a website providing comprehensive, unbiased reviews about consumer electronics; or a website that people can use to track information relating to ongoing solar flares. These kinds of websites will have a relatively consistent level of “background” usage, but every now and then, they may get a surge in usage through “word of mouth” on, e.g., reddit or other forum websites. If this happens, they may be unable to deal with the demand. However, because most of the people who visit the site will not be converted to consistent users in that situation, it may be acceptable to turn them away in favor of having consistently low costs for servers.
Neither situation applies here. Demand is unpredictable; although we are confident that demand may sharply increase, we could be wrong, and with our limited financial resources, buying digital infrastructure now in expectation of future success could be disastrous. Moreover, even if we are correct, we could not say with any certainty when demand will increase. It might be a week from now, in which case all is copacetic; it might be a month from now, in which case we’ve paid for a month of these high-powered servers that we couldn’t use them for. A month of depreciation for which we get no value, in other words. So we cannot just buy servers of our own. However, we also cannot let our server purchases lag behind our traffic, because we are trying to aggressively expand. With many or most customers, we will get one shot: the first time they visit our website, it will either work as intended, or else they will abandon it. We consequently should use a solution that lets us meet all the unpredictable demand we have without buying unreasonable amounts of digital infrastructure in advance.
