Database Reference
In-Depth Information
What does all of this, you might ask, have to do with Essbase Studio? The short answer
is: Everything. understanding a client's data is about iteration. We often overlook one
important aspect of Essbase deployments in tool selection, and actually in development
and consulting in general—the aspect of prototyping. you may never fully understand
all the nuances or complexities until you hit roadblocks in the build process. Studio is a
great tool for the prototyping phase (if used for nothing else).
It may seem strange that a book focused on application deployment should dedi-
cate thoughts and pages to a potential nonproduction activity. however, there is often
no more important phase in a deployment than developing the skeleton models, cal-
culations, rules, and data files in an effort to more fully understand both the user
requirements and the availability of the data to meet those needs. In short, prototyping
is
good development, and Studio is a great tool to help with this phase of an Essbase
implementation.
3.2 unDerstanDing this Chapter
Before we progress, it is important to understand something of the nature of this chap-
ter. It is focused tips and tricks for leveraging Studio for rapid prototype development.
That is not to say that the advice in the following pages is not useful from a production
deployment perspective (it is often equally so), only that final production decisions are a
combination of factors that, in many cases, are driven by considerations beyond the best
technical practices of a single component.
he assumption in this chapter (and much of this topic) is that the reader is already
(to some degree) familiar with Essbase and Essbase Studio. Instead of focusing on cre-
ating a flowing narrative that goes step-by-step from data source to cube, this chapter
highlights key areas of the modeling process in an effort to provide guidance for effec-
tive use of the tool.
3.2.1 Understanding the Studio Process
While this chapter relies on the reader's knowledge of Studio, it is also important to
build a framework we can use to discuss the content. to that end, we should start with
a baseline understanding of what it takes to move from data source to cube. In short,
there are five general steps:
1. map data sources (relational, text, oBIEE)
2. model the data source (create required joins, understand the data)
3. Build hierarchies and custom members
4. model the cube (create the cube model and schema)
5. Deploy to Essbase
The sections that follow directly relate to the five phases listed above. While it is not
the intention of this chapter to provide a step-by-step workflow through Studio, the tips
and tricks are organized to follow a logical progression.
3.3 mapping the Data sourCes
one of the more interesting aspects of Studio, compared to EIS (Essbase Integration
Services) and other building methodologies, is that it abstracts the complexity of data
access and treats both relational and text-based sources in much the same manner. note
