Summary of Model Options

A model can contain many elements in addition to cubes and subject areas. The additional elements are discussed in the Advanced DeepSee Modeling Guide. For reference and planning, this chapter summarizes all the elements, from both this book and the Advanced DeepSee Modeling Guide. This chapter discusses the following topics:

Items That You Can Use in Pivot Tables

This section compares items that you can use directly in the Analyzer to define a pivot table. These items are categorized as follows:

Items That Group or Filter Records

The following items can be used to group or filter records. Except for some searchable measures, none of these items can be used as measures.

Item	Use in Rows and Columns options	Use in Filters option
dimension or level	Yes	Yes
dimension or level of a related cube	Yes	Yes
shared dimension or level in the same compound cube	Yes	Yes
calculated member (non-measure)	Yes	Yes
calculated member (non-measure) in a related cube	Yes	Yes
calculated member (non-measure) in another cube in the same compound cube	No	No
computed dimension	Yes	Yes
computed dimension in a related cube	Yes	Yes
computed dimension in another cube in the same compound cube	No	No
named set	Yes	Yes
named set in another cube	No	No
named filter	No	Yes
named filter in another cube	No	No
searchable measures	Yes, except for text, string, and iKnow measures	Yes

Related cubes, compound cubes, computed dimensions, and iKnow measures are discussed in the Advanced DeepSee Modeling Guide.

Named filters are discussed in Using the DeepSee Analyzer.

Note that calculated members that are part of existing cube dimensions do not appear in filters unless the Show Calculated Members in Filters option is selected.

Items That Behave Like Measures

The following items behave like measures and are shown in the body of a pivot table. With one exception (noted), these items cannot be used for filtering.

Item	Use in Rows and Columns options	Use in Measures option	Use in Filters option
measure	Yes, except for text, string, and iKnow measures	Yes, except for text, string, and iKnow measures	Yes, if searchable
measure of a related cube	No*	No*	No
measure of another cube in the same compound cube	Yes	Yes	No
calculated measure	Yes	Yes	No
calculated measure from another cube	No	No**	No
quality measure	Yes	Yes	No
quality measure from another cube	Not applicable**	Not applicable**	Not applicable**
property of pivot-type plugin	Yes	Yes	No
pivot-type plugin from another cube	Not applicable**	Not applicable**	Not applicable**

Related cubes, compound cubes, iknow measures, quality measures, and plugins are discussed in the Advanced DeepSee Modeling Guide.

*These measures would not be aggregated correctly if used this way.

**Quality measures and plugins are designed to be directly associated with any cubes where they are to be used.

Properties

The following table summarizes how properties can be used:

Item	Use in Rows and Columns options	Use in Measures option	Use in Filters option
property	Yes	No	No
property in another cube (related or compound)	Yes	No	No

Related cubes and compound cubes are discussed in the Advanced DeepSee Modeling Guide.

Items That You Cannot Access Directly in the Analyzer

For reference, note that you cannot directly access the following items in the Analyzer:

KPIs
Term lists
Aggregate-type plugins
Ensemble business metrics

Except for business metrics, however, you can define calculated members that use these items; see the next section. Then you can use those calculated members in the Analyzer.

These items are all discussed in the Advanced DeepSee Modeling Guide.

Items That You Can Use in Calculated Members and Measures

Via calculated members and measures, you can greatly extend your model without rebuilding any cubes. The following table summarizes all the kinds of model element that you can use within the definition of a calculated member or measure:

Item	To Access This Item from a Calculated Member or Measure
MDX standard cube items (dimension, hierarchy, level, measure, property, calculated member, named set)	Many options; see Using MDX with DeepSee.
computed dimension*	Use the dimension and member name to create a member expression that refers to the desired member, in the same way that you do for any standard member.
quality measure*	Use the %QualityMeasure dimension to create a quality measure expression that refers to the measure.
KPI or plugin*	Use the %KPI function to refer to the value of a property in the KPI or plugin.
term list*	Use the %LOOKUP function to return the value of a term list item. Use the LOOKUP function to return a given field (the value field or another field), for a term list item. Use the %TERMLIST function to create a set of members, given a term list pattern.

*These items are discussed in the Advanced DeepSee Modeling Guide.

DeepSee does not provide a way to access a Ensemble business metric from within a calculated member.

Comparison of Possible Widget Data Sources

The preceding chapter introduced pivot tables, which are the most common kind of data source for a widget on a dashboard. DeepSee provides many other kinds of data sources. You can directly use any of the following items as data sources:

Pivot tables
KPIs (see the Advanced DeepSee Modeling Guide)
Pivot-type plugins (see the Advanced DeepSee Modeling Guide)
Ensemble business metrics (see Developing Ensemble Productions)

The following table compares these items:

Feature	Pivot Tables	KPIs	Plugins	Ensemble Business Metrics
How definition is stored	Folder item	Class definition	Class definition	Class definition
Where defined	Analyzer	Studio	Studio	Studio
Data source options	MDX query	MDX query SQL query Values returned by custom code	MDX query	Values returned by custom code
Supports custom computation using lowest-level records?	No	No	Yes	No
Supports filtering?	Yes	Yes, depending on your implementation	Yes, depending on your implementation	No
Supports custom filtering (for example, provide user with only a subset of the members of a level)?	No	Yes, depending on your implementation	Yes, depending on your implementation	No
Supports listings?	Yes	Yes, depending on your implementation	Yes	No
Supports the Mini Analyzer?	Yes	No	No	No

High-Level Summary of Options

For reference, the following table summaries the possible contents of a DeepSee model, including information on which tool you use to create each element:

Item	Purpose	Location of definition	Primary tool to create	Where described
dimension, hierarchy, and level	Define groups of records	cube class	Architect	“Defining Dimensions, Hierarchies, and Levels”
measure (numeric, integer, date, age, or boolean)	Aggregate values across multiple records	cube class	Architect	“Defining Measures”
measure (text or string)	Store string data in the fact table. These measures are usually also searchable.
searchable measure	Aggregate values across multiple records; also support filtering of lowest-level records
iKnow measure	Used by iKnow dimensions; not visible in the Analyzer	cube class	Architect	“Using Unstructured Data in Cubes (iKnow)” in the Advanced DeepSee Modeling Guide
property	Contain data specific to a level	cube class	Architect	“Defining Properties”
listing	Provide access to lowest-level details	cube class	Architect	“Defining Listings,” “Defining List Fields”, and “Defining Listing Groups”
calculated measure	Define a measure based on other model elements	cube class or pivot table	Architect or Analyzer	“Defining Calculated Members”
calculated member (non-measure)	Define a member based on other model elements	cube class or pivot table	Architect or Analyzer	“Defining Calculated Members”
named set	Define a reusable set of members	cube class	Architect	“Defining Named Sets”
subject area	Filter a cube or otherwise refine its definition	subject area class	Architect	“Defining Subject Areas”
computed dimension	Define groups of records to retrieve at runtime (usually via SQL or MDX)	cube class	Architect	“Defining Computed Dimensions” in the Advanced DeepSee Modeling Guide
quality measure	Define measure via MDX outside of a cube	quality measure class	Quality Measure Manager	“Defining Quality Measures” in the Advanced DeepSee Modeling Guide
related cubes	Use levels of a different cube	cube class	Architect	“Defining Cube-Cube Relationships” in the Advanced DeepSee Modeling Guide
compound cube	Combine measures from different cubes or see these measures side by side	subject area class	Architect	“Defining Shared Dimensions and Compound Cubes” in the Advanced DeepSee Modeling Guide
KPI	Query the data in a more flexible way, for use as an alternative data source	KPI class	Studio	“Defining Basic KPIs” in the Advanced DeepSee Modeling Guide
pivot-type plugin	Calculate a value based on the lowest-level data; use via drag and drop	plugin class	Studio	“Defining Plugins” in the Advanced DeepSee Modeling Guide
aggregate-type plugin	Calculate a value based on the lowest-level data; use in calculated members	plugin class	Studio	“Defining Plugins” in the Advanced DeepSee Modeling Guide
term list	Define values outside of a cube, for various purposes	DeepSee folder item	Term List manager	Defining Term Lists in the Advanced DeepSee Modeling Guide
business metric	Compute data specific to an Ensemble production	business metric class	Studio	Developing Ensemble Productions
named filter	Define a filter for use with a specific cube	global	Analyzer	Using the DeepSee Analyzer