Header DefinitionAs mentioned earlier, Essbase requires a representative of all of the dimensions in the database outline to be included in the data you are attempting to load.. While cr
Trang 1Header Definition
As mentioned earlier, Essbase requires a representative of all of the dimensions
in the database outline to be included in the data you are attempting to load
Obviously, this is not always possible or even practical Never fear, the Essbase
data load rules file once again has turned this potential issue into a non-issue!
While creating the data load rules file, one of the first things you do is map the different fields in the data to the appropriate dimensions the data will load to If even one of the dimensions is missing, you will get an error when you try to validate the Essbase data
load rules file This is where the Header Definition function comes into play.
For example, the monthly file you receive from the marketing department contains
sales information for the company The data values are always in U.S dollars, but
there is no column or field in the data that specifically says US DOLLAR like you
have defined in the database outline All you need to do is quickly and easily add
the US DOLLAR member in the Header Definition and Essbase will load data as
if every record in the file contains US DOLLAR data values
To set the Header Definitions:
1 In EAS, open the Data Prep Editor
2 Click on Options | Data Load Setting.
3 Specify US Dollar in the Header Definition tab as shown here:
Trang 2Step 8: Associating fields with Data Load
Properties
In order to correctly load the data into the Essbase cube, we need to map each field
in the data, using the data load rules file, to their specific database outline
dimensions This way, while Essbase is loading the data into the database, the data
fields are correctly aligned to the right members in the database outline Essbase
will only use the fields you have mapped in your data load rules file If the data file contains more fields than necessary, Essbase can simply ignore them if you have
them tagged as Ignore.
This is the key step while setting up the data load rules file
To open the Data Field Properties:
1 In EAS, open the Data Prep Editor.
2 Click on Fields | Properties.
3 As shown below, Data Field Properties option has three tabs: Global
Properties, Data Load Properties, and Dimension Build Properties.
Trang 3Let us talk about the Global Properties and Data Load Properties as Dimension
Build Properties were already covered in an earlier chapter
Global properties
As the term global applies to everything, in this context, global means both
the Dimension Build Properties and Data Load Properties
These are the settings that can be used in the Global Properties tab:
1 Case: There are three types of Case present: Original, Lowercase, and
UpperCase While updating the dimensions, you sometimes want all
of the members to be upper case or lower case, or it can be left as originally
sent in the data file Original is selected by default This is particularly
important when you are working with Essbase installed on a Unix platform
as it is case sensitive by default
2 Prefix or Suffix: This is to add a prefix or suffix to the member so that
the member name is unique You can add your own prefix or suffix
3 Drop leading/trailing spaces: This is used to trim the member before it
gets loaded into database outline This is selected by default
4 Convert spaces to underscores: In the earlier versions of Essbase, spaces
were not allowed Therefore, this property was used to convert spaces to
underscores This is no longer a problem in newer versions of Essbase, but
this property setting is carried over to allow upgraded systems to continue
operation without major rework
Data Load Properties
In the previous section, you learned how to set the global properties for your data
load or dimension build In this section, you will learn how to add only the data load field properties Each field in the rule file can be an outline member, data member, or
data that can be ignored In the Data Load Properties, we will define each field and
how it is related to the Essbase outline
Trang 4The settings that can be set in the Data Load Properties tab are as follows:
1 Field name: This is where you map your data element to the outline member You are currently seeing MARKET_ID, which is the column name in the
relational database where our data is coming from The outline member
name is different from this column name so you need to select the field name, clear the information that already exists, and then double-click on the correct
outline dimension You should then see the member name in the Field name
(wrapped with double quotes if you have space in the member name) If
you do not see the outline and its members, click on the Outline… button
to associate this rule to the correct Essbase database Click the OK button to complete this, or click the Next button to go to the next field.
2 Data field: The Data field can be defined in two ways, one is by defining
the field name like in step 1 If all of the data you are loading is for the
same value, Sales, then you can label this column as Sales If this data value column will represent values for more than one member, that is Sales and
Stocks, and the data source contains another column that identifies the data member, then the data values column can be named only Data
Trang 5As shown in the following screenshot, on field 7 we have defined the
Account dimension name, Total Revenue, and on field 8 we have defined
it as simply Data:
3 Ignore field during data load: For any fields that are not required to
successfully load the data from the file you have received, all you need
to do is check the Ignore field during data load option in the Data Load
Properties tab.
4 Scale: Depending on the size of your company or the size of the egos of the
company executives, you may find yourself gathering data that has been
scaled to varying degrees Because of this, Essbase offers you the option to
scale your data as you load it Typically, a busy executive may feel that the
number 10,500,000 is too large and bothersome to look at So, he will ask
that his report have the data values scaled into the thousands (1/1000th)
This would give the report the presentation value of 10,500 which would
represent 10,500,000 To do this, select the Scale option and enter the
appropriate scaling value you wish to use on the data being loaded
(10, 100, 1,000, and so on)