How to Design Optimal Infocube

Hi,
Can any one explain how to design a optimal infocube, means how many dimensional tables to be maintain, and on wht basis r how to decide wch infoobjects should be maintained in dimensional table, and whn we go for line item dim., etc.....
Regards
Ramakrishna K

Hi,
We can always change the infocube desing unless there is no data in the cube.
By desining means the the characteristics in the cube are rearranged in the dimensions to make the size of the dimension table as optimal as possible and that is done through the logical grouping of the characteristicsinside the dimensions.
Just identify the chars which are logically related that is whose combinations doesn't changes frequently.It will result in less no of records and the drilldown during the report execution will be consume less time.
It's always when you go for optimal design check or the peformace check.
One of the exapmle is the program SAP_INFOCUBE_DESIGNS which gives you the ratio of the sizes of the various dimension table.
This is used to check to make sure that the ratio of the dimension tbale size should not increares more then the total size.
In other owrds no dimension tbale should have size more then 11% of the total size.
you can see the incorrect entries in red in this program.
One you get the error for particular table then you will have to rearrange the the chars of dim table that could include removing from one dim table and adding into another untial you get accurate dim tables..
This is one of the handy exapmles there can be may more depending upon the scenarios.
Hareesh

Similar Messages

How to design infocube and ods

Hi all,
How to design infocube and ods. What are the steps involved in this process. please send me to : [email protected]
thanx in advance
bhasakr

Hi Bhaskar ,
I found the following links which is very good as it shows how to create infocube step by step from basic with snapshots.
http://www.bridgeport.edu/sed/projects/cs597/Fall_2003/vijaykse/step_by_step.htm
Also check the foll.
http://www.awprofessional.com/articles/article.asp?p=28666&seqNum=3&rl=1
http://help.sap.com/saphelp_nw04/helpdata/en/80/1a643fe07211d2acb80000e829fbfe/content.htm
http://www.crestech.org/Crestech/CourseCatalog/Acrobat/Sap/sap_bw_80hrs.pdf.
for creating ods objects
http://help.sap.com/saphelp_nw04/helpdata/en/4a/e71f39488fee0ce10000000a114084/content.htm
http://www.nesug.org/html/Proceedings/nesug04/pr/pr02.pdf.
http://64.233.167.104/search?q=cache:cg6dvsTAmxUJ:www.vdestiny.com/CourseOutlines/PDF/The%2520Output%2520Delivery%2520System%2520in%2520SAS%2520Software%2520-%2520Basics.pdfcreatingods&hl=en&gl=in&ct=clnk&cd=10
Please reward for the same.

Please help how to design query to implement this requirement.

Dear Expert,
Here is what I have and what the requirement is:
I have a InfoCube:
Dimensions:
1. Material
2. Currency
Key Figures:
1. Qty
2. Amount
We want to get the Qty and Amount by Material, but use Currency to be the filter.
It means for example, if the Currency have CNY for one material, we need to show all the Qty/Amount, not only Qty/Amount in CNY.
So please help give some idea about how to design the query to implement this requirement.
Thank you,
Andy

Hi Andy:
To accomplish your requirement you need to do 2 things:
1. Create a Variable for the Currency (to be used as the Target Currency not as a filter for the Characteristic itself).
2. Use the Standard Currency Conversion for the Amount Key Figure (Conversion Tab).
This way, while executing the report you can select any currency you want and all the amounts will be converted to that currency.
Take a look at this paper:
"How to... Use Variables for Currency Conversion"
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/287bab90-0201-0010-f48e-cc55b0cd13d0?quicklink=index&overridelayout=true
A more detailed explanation can be found on this article provided by Ramakrishna Gattikoppula:
"Currency Conversion in BI 7.0"
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/a0d5bf96-b19b-2c10-e3b6-e2f12a3de99a?quicklink=index&overridelayout=true
In summary, to work with the Standard Currency Conversion you need to define 4 things:
- Exchange Rate Type.
- Source Currency.
- Target Currency.
- Time Reference.
Regards,
Francisco Milán.
Edited by: Francisco Milan on May 31, 2010 11:19 AM

How to design ODS

Hi gurus,
How to design ODS, what are the guidelines for designing it. Is there any difference between Infocube design and ODS design. please send to : [email protected]
thanks and regards
bhaskar

Hi,
http://help.sap.com/saphelp_sem40bw/helpdata/en/e3/e60138fede083de10000009b38f8cf/frameset.htm
Check your mail box.
Assign points if helpful.

How to design Master datas?? Whats is Time Scenarios???Explain..

How to design Master datas?? Whats is Time Scenarios???
Can anybody explain about Time Scenarios???

Hi..
to tell it is huge ..
but a slice of it
Design           Create design documents
Prod.           Make the product
Quality           Confirm quality of product
Sales          - Market the product
Purchase                    Procure the items
Accounts                       To control the cost of Mfg
These steps should be understood as a general approach. To what extent they must be carried out depends
on the actual situation and the experience of the project members involved.
After deciding on the business process being dealt with, the basic steps to implementing a BI based solution
are:
1. Focus on the structure of information
Develop a complete understanding of the underlying business processes. Create an Entity Relationship
Model (ERM) of the business process
The ERM as a function of the information
2. Focus on analytical needs - Overcome model complexity
Create a valid data model. Translate the ERM to the Multi-Dimensional Model (MDM) / Star schema
The MDM as a function of the analytical processing
3. Build the solution as a part of an integrated data warehouse
The Star schema on the BI stage are the InfoCubes. Translate the MDM / Star schema to one or more
InfoCube.
coming to time ..
How real-world changes are dealt with, i.e. how the different time aspects are handled is the most
important topic with data warehouses.
The attributes of a characteristic that will reside in its master data table are determined in the modeling
phase. Each attribute can be defined individually as being time dependent:
There is one time dependent check box for each attribute in the attribute tab page section.
Time dependency of an attribute allows you to keep track on the changes over time of the relation of the
characteristic and the time dependent attribute values.
In terms of technical implementation, two master data tables exist if we have both non-time dependent
and time dependent attributes.
One master data table stores all relations to non-time dependent attributes (name of the table:
/BIC/P<InfoObject name>) and
One table stores relations to time dependent attributes (name of the table: /BIC/Q<InfoObject
name>).
The time dependent attributes master data table has additional DATETO and DATEFROM system
attributes. In queries the different constellations are addressed using the key date ( Query properties).
The validity attributes are not available for navigation.
The text table, or better the description attributes, may be defined as time dependent.
SID tables with respect to master data:
The SID table is always generated if an InfoObject is not defined as attribute only (tab page general).
This table is used if the access to an Infocube or DataStore Object uses a navigational attribute or if the
access is via a characteristic without attributes. Name of the table: /BIC/S<InfoObject name>
The non-time dependent attribute SID table of a characteristic for access via non-time dependent
attributes. Name of the table: /BIC/X<InfoObject name>
The time dependent attribute SID table of a characteristic for access via time dependent attributes.
Name of the table: /BIC/Y<InfoObject name>
with regards,
hari kv

How to Identify larger Infocubes where the dimension entry count is large

Hi SDN,
I need some help on Indexes dropping and Recreation.
Please can any body help me on How to Identify larger Infocubes where the dimension entry count is as large as the Fact Table count.
Need to suggest on a solution to implement the Indexes creation for optimizing the load performance of Larger Infocubes... Its urgenTT!!!!!!!
Please help!!
Thanks,
MH

Hi,
I have ran the Report in SE38, also I have downloaded it to local Excel file.
It is showing density, rows etc.Please can you elaborate on how to easily Identify
the Larger cubes. As I have seen that the rows of both E and F fact table are displayed. How can I use this reports information in order to suggest for the Indexes?
Help in this regard is much appreciated!
MH

How to crete transactional infocube

Hi guys,
Can any body tell me how to create transactional infocube.what is difference between transactional infocube and basic infocube and in what situation we will use transactional infocube.
full points will rewarded for good answer

Hi,
how to create transactional infocube?
Click on create infocube---then u will have options basic ,transactional .select option transactional .
what is difference between transactional infocube and basic infocube and in what situation we will use transactional infocube?
Transactional InfoCubes differ from Basic InfoCubes in their ability to support parallel write accesses which means data can be written to the infocube by several users at the same time . Basic InfoCubes are technically optimized for read accesses only.
Transactional infocubes are used for planning purposes.Basic cubes are not used for planning purposes.

From a Technical Architect point of view, how to design a web application?

Hi to all,
I am writing this issue here because I this question is not related on how to program, but on how to design a web application. That is, this question is not related to how to program with the technologies in J2EE, but how to use them correctly to achieve a desired outcome.
Basically I know how to develop a simple web application. I like to use SpringFramework with AcegiSecurity and Hibernate Framework to persist my data. I usually use JBoss to deploy my web applications, however I could easily use Tomcat or Jetty (Since I am not using EJB�s).
However I have no idea on how to develop (or a better word would be �design�) a website which is divided into different modules.
Usually when you develop a website, you have certain areas that are public, and other areas that are not. For example, a company would want anyone on the Internet to download information about the products they are selling, however they would only want their employees to download certain restricted information.
Personally I try to categorise this scenario in to common words; Extranet and Intranet.
But � (and here starts the confusion in my mind) should I treat these two as two projects, or as one project? The content to be displayed on the Extranet is much different then the content to be displayed on the Intranet and definitely clients should not be allowed to the Intranet.
First approach would be to treat them as the same project. This would be perfect, since if the company (one day) decides to change the layout of the website, then the design would change for both the Intranet and the Extranet version. Also the system has a common login screen, that is I would only need to have employees to have a certain Role so that they have access to the intranet, while clients would not have a certain Role and thus they would not be allowed in. But what about performance and scalability? What if the Intranet and Extranet have to be deployed on the different Hardware!?
The second approach is to threat them as two separate projects. To keep the same layout you just copy & paste the layout from one project to another. However we would not want to have two different databases to store our users, one for the employees and the other one for the clients. Therefore we would build a CAS server for authentication purposes. Both the Intranet and the Extranet would use the same CAS server to login however they could be deployed on different hardware. However what if we want to change the design. Do we really want to have to just copy and paste elements from one project to another? We all know how these things finish! �We do not have time for that � just change the Extranet and leave the Intranet as it is!�
The third approach (and this is the one I like most) is to have a single project built into different WAR files. The common elements would be placed in all WAR files. However in development you would only need to change once and the effects would show in the different war files. The problem with this approach is that the project will be very big. Also, you will still need to define configuration files for each one of them (2 Web.config, 2 Spring-Servlet.config, 2 acegi-security.config, etc).
Basically I would like something in the middle of approach 2 and approach 3! However I can identify what this approach is. Also I can not understand if there is even a better approach then these three! I always keep in mind that there can always be more then two modules (that is not only Intranet and Extranet).
Anyways, it is already too long this post. Any comments are more then welcome and appreciated.
Thanks & Regards,
Sim085

Hi to all,
First of all thanks for the interest shown and for the replies. I do know what MVC or Multi-layered design is and I develop all my websites in that way.
Basically I have read a lot of books about Domain-Driven Design. I divide my web applications into 4 layers. The first layer is the presentation layer. Then I have the Facade layer and after that I have a Service layer (Sometimes I join these two together if the web application is small enough). Finally I have the Data Access layer where lately I use Hibernate to persist my object in the database.
I do not have problems to understand how layering a web application works and why it is required. My problem is how to design and develop web applications with different concerns that use same resources. Let me give an example:
Imagine a Supermarket. The owner of the Supermarket want to sell products from the website, however he wants to also be able to insert new products from the website itself. This means that we have two different websites that make use of the same resources.
The first website is for the Supermarket clients. The clients can create an account. Then they can view products and order them. From the above description we can see that in our domain model we will have definitely an object Account and an object Product (I am not mentioning all of them). In the Data Access layer we will have repository objects that will be used to persist the Account and Products.
The second website is for the Supermarket employees. The employees still need to have an account. Also there is still a product object. This means that Account and Product objects are common to the two websites.
Also important to mention is the style (CSS). The Supermarket owner would like to have the same style for both websites.
Now I would not like to just copy & paste the objects and elements that are common to both websites into the two different projects since this would mean that I have to always repeat the changes I make in one website, inside the other one.
Having a single WAR file with both websites is not an option either because I would not like to have to deploy both websites on the same server because of performance and scalability issues.
Basically so far I have tought of putting the common elements in a Jar File which would be found on the two different servers. However I am not sure if this is the best approach.
As you can see my problem is not about layering. I know what layering is and agree with both of you on its importance.
My question is: What is the best approach to have the same resources available for different websites? This includes Class Files, CSS Files, JavaScript Files, etc.
Thanks & Regards,
Sim085

How to design this report ?

Hi experts,
How to design the customer total outstanding report as on a particular date ?
We also want Debit amount and Credit amount in the query output..
Regards,
Nishuv V.

HI,
if u want the current date u go for the customer exit (that means daily)if he wants aparticular date then u go for the user entry variable in 0calday,in variable screen he sould mention the date for that date only the out standing report will come .0calday and customer u keep in rows and those debit and credit u keep in the columns.
if daily as on date they want outstanding u go for the customer exit just u create the variable on 0calday with customer exit if variable name is 'zcedate' go to the tcode CMOD and write the below mention code.
when 'zcedate'.
clear l_s_range.
l_s_range-sign = 'i'.
l_s_range-opt = 'eq'.
l_s_range-low = sy-datum.
append l_s_range to e_t_range.
Thanks for giving this opportunity.
Thanks & Regards
k.sathish

How to design the report?

Hi,
how to change the filling data in report by vertical way?
for eg: i am creating a group (group by userId) and displaying user name and access rights info;
User Name   Access Rights
TestUser      Adminuser
TempUser    NotanAdminuser
but i want in this format
UserName        TestUser   TempUser
AccessRights Adminuser NotAnAdminuser
how to design the report?

i haven't tried using cross tab.
but my requirement is like...
Modified User Name ........ TestUser     Thendral    till 'n' number of users
Total NO of Patients..........10                    5
          Heart Failure...........5                      1
          Surgical..................1                      2
          Pneumonia             3                      1
          Chest Pain             1                       1
        till 15 rows
How to achieve this using crosstab?

How to design a 3D database in SDO

Hi,
I need to design a 3D database in Oracle9i SDO. As we know, this SDO support the 2D GIS seamlessly. How to design a 3D database in Oracle9i SDO,
Could any person have some experiences. Thanks!
Best regards!

Hello All!
Due to my diplom thesys I have to create a real 3D (not 2 1/2D) database in Oracle. You seem to have a little experience in this subject and I hope you can tell me whether my plan is possible or not...
The data I'd like to store in the database is/are BRep, means a 3D-Objekt composed of many coplanar 2D-planes. I think storing the data won't be the problem because I can store the vertices of the 2d-polygons with x,y,z values. The queries will be the main problem. I have to query all types of intersection, neighborhood and some distance between the objects.
I know, SDO_RELATE is the only filter which supports three dimensions, but there's something I don't understand:In the spatial user's guide it says:
"the SDO_RELATE operator, can be used to determine with certainty if objects interact spatially".
WHAT does this mean in case of 3D? True if the two objects intersect, false if not?
Hongwei: How is your project doing? maybe we can exchange some experience...
Many Questions, hoping for some help...;-)
regards from germany,
Markus Reuter

How to design HTML Table in ADF

I am new to ADF Tech,
I would like to know, how to design the HTML Table Rows and Columns in ADF
Ex:
<TABLE width="100%" border="1">
<TR>
     <TD>GUID</TD>
     <TD>123</TD>
     <TD>Name</TD>
     <TD>Mark Antony</TD>
     <TD>Version</TD>
     <TD>1.0</TD>
</TR>
<TR>
<TD>Created</TD>
     <TD>Oracle</TD>
     <TD>Modified</TD>
     <TD>Oracle User</TD>
     <TD>Placements</TD>
     <TD>20</TD>
</TR>
</TABLE>
Thanks in Advance

Balaji,
With JSF in general (and ADF Faces too), you should not think in terms of HTML output, but in terms of JSF components. ADF has an af:table component that renders things in rows-and-columns, but emits HTML that is much more complex than just a simple HTML table.
John

How to design templates with Microsoft Excel

Hi
When creating a new template in EBS, we have "Microsoft Excel" as one of the the template types (just like RTF). How to design the template if Excel is selected as the template type. I uploaded a blank EXCEL file and the concurrent program errors out with "Excel Processing" exception.

hi,
r u uploading the blank excel..r the excel u got from the Analyzer for Excel from the BIpublisher.
One small clarification on my previous post..like the .xdo not the excel name.Inside the excel it is the sheet name.
If u uploading the excel that u got from the Analyzer for excel means it should work.
Plz select the template type as Excel after uploading the template.
Its working for me.
This link will be helpful for u
http://blogs.oracle.com/xmlpublisher/2007/05/16/
Edited by: Ananth.v on Mar 4, 2010 3:07 PM

How to design a particular screen?

Hi All,
In my apllication i am working on oracle forms using manually. I would to design a
particular page with the following attributes as shown below?
How to design sunch kind of pages.
Column names
Expir
Reocrd Year Mnt Ct Mnt Year Fuel Fuel code Mine type Reported fips source name
Type
{select list} {Text item} {text Item} {Select list} {Text Box} {Text box} {Text box} {select list} {select list} {text item} {select item} {text item}
till Mine type i do not have any scrolling type, But from reported onwards a scroll should apper so that i can scroll to right and seen what are the attributes
still exists.
Since this is a form, how can we maintain or develop the screen.
Hope u have understood my problem.
Or can anybody give let me know how to display or attache my screen shot, so that u may clear idea on this.
Thanks,
Anoo..
Edited by: Anoo on Mar 2, 2010 4:38 AM

Hi Ben,
How do we use tabular form can u let me know the procedure to design it? bec i have not aware of that.
-Anoo..
Edited by: Anoo on Mar 2, 2010 4:41 AM

How to design EDW for source systems from different Time-Zones

How to design EDW for source systems from different Time-Zones?
Suppose IT landscape has a global BW in New York, and source systems in americas, europe and asia, then how the time-zones effect on time related things like delta selections on date or timestamp etc.

As you said BW is global in NY, your source system must be global too. People from various locations can connect to same source system and thus timestamps for delta is always maintained as 1 single time. We have same scenario in our project. Our R/3 system is used by users in US and Europe. So we run deltas twice in day to make sure we got deltas from both locations.
If scenarios was such that all locations connect to separate R/3 system, then obviously you have multiple queues. That is unique delta queue for each source system so deltas will be pulled as per data in respective queues.
Abhijit

How to Design Optimal Infocube

Similar Messages

Maybe you are looking for