Business Intelligence and the Cloud: Strategic Implementation Guide

CHAPTER 3

Deployment Models

The IT/IS landscape is changing. For many years we deployed IT/IS in the local data center and on the user desktop. However, we have seen vast evolutionary changes in how IT is delivered to users. Some say that we are returning to the way of the mainframe, where all of the processing is occurring in the data center and the user has a dumb terminal on his or her desk that does not do much more than display the results that a mainframe spews out. That is simply not true. We are doing much of the processing in locations other than the desktop; large transaction processing systems like book buying at Amazon or large database systems like the management of customer relationships at Salesforce are doing the processing, and they are delivering their results to the user to display. Following are some major differences between mainframe and cloud processing:

• The digital connection that organizations have to other users and services from cloud providers that is enabled by the Internet.
• The shared processing that occurs between the desktop and the cloud provider.
• Changes in deployment of IT that affect the cost structure of operational and capital expenses (this might have the greatest effect on innovation and deployment).

Throughout this book, the terms a cloud and the cloud are used somewhat interchangeably. That is a bit at odds to what is often discussed in marketing literature as “moving to the cloud.” That phrase is a bit of a misnomer, but it sometimes ...