Hyperion Interview Questions and Answers

1.Q:What is the meaning of the two options from Essbase regarding storage and what separates them?

A: a)The Aggregate Storage Option or ASO:1. Used for more than ten dimensions 2. The majority of the dimensions will be sparse 3. It is impossible to write back in ASO. In the next versions we should be able to write back.

b)The Block Storage Option or BSO:1.Less dimensions and more dense dimensions

2.The write back quality is present, good for application planning. The commercial programs are usually dense, the analytical cubes-sparse. We must comprehend the meaning of the sparse and dense members. Dense means that the cells of two dimensions  have intersections that include a value, let’s have this example of two dimensions like period (Month) and region(City, Country)and we have to calculate how much sales are. Usually the biggest regions will get a specific value for a month. We can talk about the dimension period as being dense. Let’s take another dimension, for example-products. The same, some of the products are being sold in a specific region, they will not have any value in other region and they will become sparse. The question that has to asked is what meaning has naming them sparse or dense? Is it any real difference? The answer can be found only by doing good research.

2.Q: Is it possible to construct dimensions straight from the data sorce and not to use the rule files?

A: No, it isn’t.

3.Q: In general why do we have to use rules files?

A: For making dimensions containing thousands of members in an automated way.

4.Q: What are the first 3 ways of constructing dimensions?

A: a)the level references.

b)the Parent/Child references.

c)generation references.

5.Q: Can we define in the generation build method generation 1?

A: No we can’t , because it’s not a valid operation.

6.Q:Can we construct a dimension in which we have added Generation 2, then Generation 4 and the following will be Generation 3 but not now, it will be added later?

A: We can’t do it in this way, if 2 and 4 are already applied we must aply Generation 3 as well.

7.Q: With the aid of a single build rule file can we make at least 2 dimensions?

A: yes, but a different rule file should be used for every dimension.

8.Q: What is the meaning of user defined attributes or UDA and what is the difference between them and Aliases?

A: The user defined attributes are referring to the members classes while the aliases are only different ways to name memebres, they are totally separate and have totally separate usages.

9.Q: Is it possible to query a member for UDA in calculating scripts?

A: Yes, we can query members for UDA in calculating scripts.

10.Q: What impact has UDA on the size of a database?

A: Because UDA’s are not needing any space in addition, they don’t have any impact on the size of a database.

11.Q:What separates Attribute dimensions from UDA’s?

A: Attribute dimensions are different from UDA’s because of their improved flexibility. The calculation with the attribute calculations dimensions with five members named: sum, min, max, count and avg is made automatically and dunamicaly.

12.Q:What impact have UDA’s  and Attribute dimensions on the performance of batch calculations?

A: The UDA’s don’t have any impact because they don’t make inherent calculations and the Attribute dimensions the same have no impact because they just make dynamic calculations.

13.Q: In what way can we show UDA’s in the reports and what impact have they on the performance of the reports?

A: We can’t show UDA’s in the reports so there is no performance change.

14.Q: What impact do Attribute dimensions have on the performance of reports?

A: The impact of the Attribute dimensions on the performance of the reports is very high due to the dynamic calculation made referencing the reports. If the number of attribute dimensions shown in the reports is high, this will have a negative impact on performance.

15.Q: In the data loading period, what is the result if both the selection and the rejection criterias are implemented?

A: The result will be the rejection of the record.

16.Q:In what way is data storage made in the database of ESSbase?

A: In Essbase (a database based on files)the data storage is made in 2GB PAG files that have a sequential way of growing.

17.Q:Is it possible to keep several databases in only one application?

A: It is possible but the generic recommendation is to use a single database per application.

18.Q: Is it possible to include a BSO and an ASO database, both in only one application and why?

A: The answer is no because BSO and ASO are defined at the making of the application (not the database). The database included by the application will be only ASO if the application is ASO.

19.Q: What file extensions are used for rule file, calc script or outline?

A: The extensions are: .RUL, .CSC and .OTL.

20.Q: What purpose do provider services have?

A:They have the purpose of communication of the Essbase with the office tools of Microsoft.

21.Q: How can we create rule file, load data or metaoutline in a different way?

A: We can make it using Integration services and the ESSbase studio from the version 11.

22.Q: Is it possible to begin and end an application at the individual level and can this enhance the performance?

A: Yes, it is. We can begin only those applications that have high traffic, in this way we can save server resources, the memory will be filled with databases at the beginning of the application.

23.Q: If we already made an application in Unicode mode , can we modify it later to become non-unicode?

A: The answer will be no.

24.Q:In what way can migrating of an application from the test environment be made if we want to move it in the production environment, also can we do this from Windows to Unix or if the test environment is different from the production environment?

A: There is  a tool given to us by the Hyperion Administrative services  but this only helps with the migration of the application and the objects database while the data is not migrating. In the other two cases the answer is yes.

25.Q:Can we run on the same machine two instances of the same Essbase server and each one will have the test environment and the development environment?

A:We will respond with yes, because we can have two instances of Essbase with various sets of services on the same machine.

26.Q: If we have A as the dimension and B,C as members of it how can we stop b and c from rolling up to A?

A: We can do it with (~)exclude of the consolidation operator.

27.Q:What is the role of the never consolidate operator(^)?

A: It stops member from becoming consolidate on all dimensions.

28.Q: Is it possible to have several metaoutlines in the Integration services that are built on a single OLAP model?

A: The answer is yes.

29.Q: Is it possible to have a metaoutline built on two separate OLAP models?

A: No, it’s not possible.

30.Q:What is the true meaning of a hybrid analysis?

A: The members with a higher level along with the data associated will stay in the Essbase database while the members of a smaller level and their associated data will live in the relational database.

31.Q:What is the explanation for a higher efficiency of a bottom-up calculation than a top-down calculation?

A: The reason for this is calculating a higher amount of blocks than needed and for the right results to show up there is a need for making a top-down calculation.

32.Q: What is the real base for calling a parallel or a serial method for making the calculations?

A: We will use the serial calculation only when we have one processor and when we have more than one the operation can be split into threads that will work on the various available processors.

33.Q:What is the meaning of a block locking system?

A: The Essbase Services, also called analytic will make a lock to each of the blocks that encapsulate the childs of the initial block, this is what a block locking system means.

34.Q: What partitioning options do we have in Essbase?

A: The  partitioning options are: replicated, transparent and linked.

35.Q: Is it true that the retrieval time is shorten by the dynamic calculation  and the database calculation time is longer?

A: It is exactly backwards than stated because the when the calculation of dynamic members is made we will have a longer retrieval time.

Few Basic Hyperion interview questions for jobs:

hyperion planning interview questions and answers

Q 1. Define Hyperion?

Ans. Hyperion is Business Intelligence (BI) and Business Performance Management (BPM ) tool. It is the market leader in Operational , Financial and Strategic Planning. It contains the applications for reporting, Planning , dashboards, Analysis,  scorecarding, consolidation, Workspace, Master Data Management and Foundation.

Q 2. Explain Olap and mention if it is related to HYPERION FINANCIAL MANAGEMENT ?

Ans. Hyperion HFM/ Hyperion Planning both are Essbase based. They are front-end tech and Ess-base is the back end. Ess-base is a M-OLAP. There are three types of OLAP technology present in  the market. These are ROLAP, MOLAP and HOLAP.

An example of ROLAP is BO. Here we deal with table and they acts as a virtual cube. But if we  think about Oracle Express Hyperion Essbase and Cognos then they are real cubes. Thus  MOLAP. ROLAP+MOLAP=HOLAP.

In case of BO, we need to join to attain the OLAP effect virtually but behind is a joined table. But in Case of MOLAP there is no concept of table it is cube only.

For example a page is a table then a Book is a Cube.

If the data is large then it is advisable to take up ROLAP not MOLAP. because the performance of the cubes degrade. If the size of data is arround 100 GB to 150 GB then it is fine to go with MOLAP and if it is more then that then we should choose BO.

Q 3. Explain why we use Hyperion?

Ans.  We have IIS for HFM although there is no IIS for OLAp’s. If there is no IIS then it is a time taking process to retrieve financial data.

Q 4. Is it possible to have one ASO database and one BSO database in a single application. Justify .

Ans. No because ASO/BSO classification is defined at Application level and not at the database.

Q 5. Can we have multiple databases in single application?

Ans. Yes, in this one database per application.

Q 6.Can we start and stop an application individually and how to increase performance using this?

Ans. We can manage our server resources by starting only applications which receive heavy user traffic. When an application is started, memory is accompanied by all associated data bases.

Q 7. Explain custom based macro?

Ans. We write Ess base calculator functions and special macro functions under Ess base macros. Custom defined macros makes use of an internal Ess base macro language that enables to combine calculation functions and also helpful to operate on multiple input parameters.

Q 8. Explain data file cache?

Ans. It represents buffer in memory which holds compressed data files (.PAG ).

Q 9. What does “DOU” mean?

Ans. In reference to RPG, DOU means (Do Until).This will execute a loop matched with an End or End do

For instance

Eval X=1


Eval X=X=1


This example will go through the loop 4 times with the result of X being 4.

“DOU” is similar to Do While(“DOW”) where the difference lies in that the DOU will always perform 1 pass through the loop and DOW will perform the comparison and only continue into the loop if the condition meets else the program continues after the End (EndDo).

For example

Eval X=1




The result is that the loop will not execute and the value of X will be 1.

Q 10. Which property helps us to consider using ACE ?

Ans.  A very little population of intercompany data forces us to consider using ACE.

Q 11. Which two functions can produce a report that includes only Elimination entities?

Ans.  two functions which can produce a report that includes only Elimination entities are a Fixed Name List and Dynamic Name List.

Q 12. Explain Dense and Sparse Dimensions?

Ans.  Dense dimension is a dimension in which most data exists for every combination of dimension members whereas sparse dimensions are the dimensions which has low probability that data will exist for every combination of dimension members.

Q 13. What are the three primary build methods for building dimensions?

Ans. The following are the three primary build methods for building dimensions :-

1)  Generation references

2) Level references

3) Parent-Child references.

Q 14. Differentiate between ASO & BSO?

Ans. 1) We cannot write back in ASO although we can write back in BSO.

2) Most dimensions in ASO are sparse whereas in BSO most of them are dense.

3) We cannot create more than 1 database in ASO but we can create more than 1 db per application. If we have more than 10 dimensions then we should opt for ASO.

Q 15. Explain attributes?

Ans. Classification of a member in a dimension is known as an attribute. We can do selection of group members based on their associated attributes. We can also specify an attribute while performing calculations and along with we can use calculation functions. As example, the db in Sample Basic with product dimension has some attributes like package type, size, and flavor. We can add attributes to the dimensions and can retrieve data. For example, to retrieve “coke with 8 Oz with bottles”.

Leave a Reply

Your email address will not be published.