The importance of time Cognos TM1 applications are most often used to view, compare, and predict information over time. For the most part, information in TM1 will be:
Take the Intellipaat Cognos training course for a rapid career growth.
When considering the concept of time in a TM1 cube you need to determine:
The lowest level of time detail required
Additionally it is important to consider:
You must consider the following points:
Granularity—the lowest level of time detail required
Granularity is the extent to which a system is broken down into small parts, either the system itself or its description or observation. The most common granularities for a time dimension are:
Month—two time dimensions
Cognos TM1 cubes that store data at the monthly level of detail may have one or two time dimensions.
Some of the advantages of separating the year and month into two dimensions include:
Disadvantages might include:
You would set up the month dimension with elements running from July through June (as July, August, September, October, November, December, January, February, March, April, May, and June):
The month dimension may also include consolidated elements for:
For non-financial cubes, the year and month dimensions may follow the structure as shown:-
For financial cubes, it may be more appropriate that the year and month dimensions use a calendar year (with a set of months running from January to December):
Month—one time dimension
In some situations, using a single time dimension may be more useful. In fact, if forecasting is the purpose of the application, you will almost always see a single time dimension:
Single time dimensions usually:
Date format flexibility
One of the most commonly requested requirements for Cognos TM1 dates is the ability to select and display the date in a variety of formats. The planning sample model which is part of the Cognos TM1 installation offers a simple example of the use of element attributes to manipulate the value of dates. The plan_time dimension has the following attributes defined as type A (alias):
Time dimension aggregation examples
Time dimension maintenance
Time dimensions, like any other dimension in IBM Cognos TM1, must first be initialized (created) and then maintained to keep the current dimension with the model’s requirements.
In some cases, the initialization of the model will include multiple forward years so that the time dimension maintenance will almost be non-existent for the foreseeable future. This means that the time dimension will include multiple dates past the current and next calendar and/or fiscal years. This is referred to as future proofing the time dimension.
Depending upon the size of the application and/or number of users (and other factors), it may be completely acceptable for a TM1 Administrator to manually modify this dimension using the TM1 Dimension Editor.
In other situations, a TI process may accept appropriate parameters and then update the time dimension.
In any case, all applications deserve a documented and tested approach to maintain the model’s time dimension.
Date and time Rules functions
The following Cognos TM1 functions deal with dates and times and therefore, I believe that they are important enough to warrant a quick review:
The DATE function returns a date string like YY-MM-DD for a given serial number. Good to know because the function NOW will return us a serial number!
# — will return a string of ‘YYYY-MM-DD”
DATES is a little more practical (in my opinion). It takes a numeric value for a given year, month, and day and returns a string in a similar format:
# — will return a string of “YYYY-MM-DD”
DAY returns a numeric value for the day in a given date string:
# — will return 25
DAYNO is reverse of DAY, the DAYNO function returns the serial date number corresponding to a given date string:
# — will return 13947.
To easily pull the month from a date string, the MONTH function returns a numeric value for the month in a given date string:
# — will return 3
This function can be used to return the current date and time from the system clock. Keep in mind though, that it returns the date/time in a serial number format.
TIME is a “sister” to NOW, the TIME function returns a string, in HH:MM format, representing the system time on the TM1 Server:
The TIMST function takes a date/time serial number and returns a formatted date/ time string based upon the format parameter’s value which you provide. You can also pass an optional parameter called extended years:
FOO=TIMST(NOW, 'M D, Y');
# — will return “DEC 27, 2011”
The TIMVL function is a lot like the TIMST function. It accepts a date/time serial number and two parameters, but returns the numeric value of a component (year, month, and so on) of a date/time value.
# — will return “2011”
This function has an optional parameter to define the resulting format. No parameter given will return a two digit year within the string (YY-MM-DD) and a parameter of 1 will return a four digit year within the string (YYYY-MM-DD):
# — will return “11-12-27”
The YEAR function takes a date string and returns the numeric year in the string:
# — will return 12Previous Next
Download Interview Questions asked by top MNCs in 2019?