Multi-User Environment for Siebel Analytics/OBIEE by Alok Chowdhary on March 13, 2009 in Siebel Street

By default, only one user can edit the repository at a time, but a more efficient environment would allow developers to modify the repository simultaneously and then check in changes. Oracle BI allows multiple developers to work on objects from the same repository during group development of Oracle BI applications.

Steps for configuring Oracle BI multi-user development environment:

1)Create Project

In the Admin tool, open the Project Manager

Path:-Select Manage > Projects and then Action > New project for creating new project.

From the figure down below, you can notice two parts, in the left you can see objects that are available for the project and the right part consists of objects that can be added in projects. Select the objects from the left part that you want to add to the project and click the Add button. If you have selected the presentation catalog, all fact and dependent objects are selected in the project.

Besides the catalog, other objects such as USER, groups, Variables and initialization blocks can also be added in projects. Apart from this, you can also remove unwanted objects from project by clicking the Remove button.

2) Set up a shared network directory

Administrator needs to identify or create a shared network directory that can be accessible to developers to keep the repository file at that location. This repository is the master repository which is accessible to multiple developers to check in or check out the changes done. Developer has to point to this directory path when they use Admin tool at their machine.

Making changes in the Admin tool at the local machine to use as a multi-user development environment:

1)Point to multi-user directory:

Set up for Admin tool to point multi-user  development directory.

Path:-Select Tools > Options and then select Multiuser tab.

From the figure,  it can be seen that it is a two field, one multi-user development directory in which you have to browse the path of the shared directory where the original repository has been kept for development purposes. The other field full name  is optional,  but if the user mentions name in that field, it helps in tracking the changes made by each user and stored in the HKEY_CURRENT_USER part of registry and is therefore unique for each login on computer.

2. Check out project:-

After pointing to the multi-user development , the directory developer can check out desired projects.

To check out projects, go to path File > Multiuser > Checkout which will be only available when the multi-user environment is setup. After this, the developer is presented with a dialog box to select the master repository if one has more than one repository. Select the repository , then enter user name and password, it will navigate to select the project or projects to be imported. After selecting the projects,  user must enter the name of the new repository which will be stored in user’s local directory.

3.Admin tool task during checkout:-

During checkout , the admin tool performs the following task:-

• Makes a temporary copy of the master repository on the local machine.
• Saves local copy of projects in the new repository on the local machine.
• Saves second local copy of project in the new repository on the l
  local machine with prefix as “original”.
• Deletes temporary copy of the master repository from the local machine.

4. Changes done in metadata:-

Changes can be performed on logical tables, table definitions and logical table sources. Developers can work on the same project but if one developer deletes objects,  it will be migrated without any warning. So developers should keep in mind that modifications can affect others too.

5. Tasks done during check-in:-

The Admin tool perform the process of locking the master repository to prevent other developers from attempting to merge at the same time and copies the master repository to a local directory so that the developer will be merging with the latest or recent repository.

6. Check in changes:-

After performing modifications on the repository, the developer needs to check in changes and merge with the master repository in the shared path. Only one developer at a time can merge it. After selecting File > Multi-user >Merge local changes, the developer is shown a  dialog box having full name and the option to write comments if any and after clicking ok , the admin tool performs the process of copying master from shared and keeps it on the local machine.

After developers lock the master repository,  the merge process take place. After the merge process,  developers have to publish to the network.  Go to path File >Multi-user >Publish to Network to publish changes done in the repository.  This will finally merge local repository changes to the master repository and at the same time a local copy of the repository has been removed from the local machine.

Advantages:

• History menu option:-This option gives the detail of all the changes performed during the merge process. We can have the version history that tracks all the changes performed in the repository during the merge process.
• It helps to track the Project history.

Disadvantage:

• Multiuser develop environment is purely for repository development. Dashboards/Reports can be developed/managed by any no of users in the browser once the presentation server is up.

Tp know more about OBIEE 

Dynamic MTD / YTD Measures in Siebel Analytics

Definition:

MTD: Acronym for Month to Date. MTD value of a certain measure would be the aggregated total of the measure from the start of the month till the defined date. For example: For the date 15th August 2008, MTD Dollars would be the total dollars from starting of August 2008 i.e. 1st August 2008 to 15th August 2008.

YTD: Acronym for Year to Date. YTD value of a certain measure would be the aggregated total of the measure from the start of the year till the defined date. For example: For the date 15th August 2008, YTD Dollars would be the total dollars from starting of 2008 i.e. 1st January 2008 to 15th August 2008.

Situation:

The client required MTD YTD metrics for a few reports. The twist in the requirement was the calculation based on user specified dates supplied from the dashboard prompt instead of the usual current date MTD YTD metrics. This means capturing the date value entered by the user and applying its relation to the concerned request.

Solution:

After a lot of brainstorming and trying out N number of unsuccessful RPD and Answers permutations and combinations, the solution was arrived at with the help of “Filter based on other request” option in the Answers section and a dummy request to capture the prompted date.

Implementation:

Step1:  Creating the dummy report

This dummy request is used to capture the Transaction date. As seen, the Transaction Date column is prompted and the dashboard prompt date value is fed to this request. We also take Month and Year columns for MTD YTD calculation purposes.

Step2: Creating the MTD Request

The request shows Amount by Pay Type. We want to calculate MTD amount for a date specified by the user. The Date filter columns of the dummy and the main request should be same.

Step3: Creating MTD specific filters

Using the “Filter based on Another Request”, we specify the filters for MTD calculation.

Step4: Creating the dashboard prompt

It’s always a good practice to specify a default value to the dashboard prompt so that the dummy request is initialized to some value. Needless to say, the dashboard prompt, the dummy request and the main request should come from the same subject area.

Step5: Displaying and verifying the MTD metrics

The MTD request has been placed on the dashboard along with the normal reporting date request. This way one can verify the MTD results quite efficiently.

Conclusion:

• The YTD metrics can be calculated in a similar manner. The month filter won’t be applicable for YTD. Rest remains unchanged.
• Combine request feature of Answers should be used if the reporting date, YTD and MTD metrics need to be shown together in one single request.
• This dummy request feature is an excellent way of capturing the user entered prompt values of any data type and manipulating the report metrics on the fly based on the captured prompt value in the dummy request.

To summarize, via this method, we get the power of the presentation variables of OBIEE, strictly to just some extent, in Siebel Analytics.

To know more about Siebel Analytics

BI Publisher Highlighted

I want a single application to print my reports, invoices, checks, labels etc. and also want the same application to send these to different destinations like printer, fax and e-mail. These reports should contain:

• Graphs
• Tables in highly formatted fashion like table headers should be repeated on each page
• Section totals
• Page totals
• Should have headers and footers
• Should have conditional formatting
• Cross Tab Support
• Easy to create templates

These are my requirements for reporting. All of these can be fulfilled very easily in Analytics dashboards, but, when it comes to printing, dashboard is not a very good choice. For these reports BI Publisher is a perfect solution. Oracle BI Publisher is a Standalone and Platform Independent reporting tool which creates highly formatted printable reports. Oracle BI Publisher uses Template for formatting purpose .Templates can be created in MSWORD, ACROBAT, EXCEL and many other Formatting Tools.
It takes the data from the source, formats it suitably using Template and prints it in different formats such as PDF, EXCEL and HTML.

The Formatted Reports can be published over Printer, Fax or Internet.Oracle BI Publisher overcomes the shortcomings of Siebel Analytics when it comes to Printing of a Report. It provides a unified solution to the above mentioned requirements, hence reduces cost, complexity and manpower.

It can use data from any source which provides JDBC connection like HTTP XML feeds, web services, file data sources. It is pre integrated with OBIEE Suit. You can also use requests developed in OBIEE as the sources of data for BI Publisher.

Data sources for BI Publisher.
BI Publisher can have many sources like OBIEE, Oracle Database, SQL Server, XML and various other data sources are supported by BI Publisher.

• SQL Query: A connection to a database has to be specified for SQL Query. Then a query to retrieve the required data is fired on the database. Bi Publisher also provides a feature called Query Builder.
• BI Answers: It uses Dashboard requests as the data source.
• Web service: It uses a web service which returns data as its data source.

Creating report templates in BI Publisher
Report templates can be created in RTF Format as well as PDF format. To create templates in RTF format you need to have any text editor with RTF support like MS Word, for PDF template, you need to have any tool which gives facility of PDF creation like Adobe Acrobat Professional.On installation, the BI Publisher Template Builder integrates itself with MS Word.

To create a simple RTF template, load the XML containing your sample data by clicking Data->Load XML Data on Template builder toolbar. Once the data is loaded, you can put the data fields in the template. You can provide groupings, conditional formatting and conditional regions. We can also use the wizard to create tables, crosstabs, and charts. Preview of the report is possible with sample data available in PDF, RTF, PPT, HTML and Excel formats. After creating these templates, these templates have to be uploaded to the server to be used by the report. The output format of reports can also be restricted; we can tell the BI Publisher to output report as only PDF and no other format.

Other features of BI Publisher

• Provides a facility of LOV (List of Values) so that data in a report can be filtered according the value selected in LOV.
• Different layouts for same report are supported.
• Bursting, split a report based on a key in the report data and deliver a report based on the second key in the report data.

– Inputs from Anand M

Read More about BI Publisher

Analyzing Analytics

Dashboard Power
The other day, I was ‘Googling’ for a Siebel Analytics document when I stumbled upon this article which is quite a good example of how the prowess of Analytics is making life easier for commuters. Siebel Analytics now rechristened Oracle BI, helped officials at Manchester airport to cope with the extra pressure of screening additional bags by identifying future trends.

It provided Manchester Airport’s management team with interactive dashboards of flight, passenger and baggage movements. As a result, Manchester Airport lifted the ‘One Hand Baggage Rule’. The full article can be found out at the following link: http://www.computerweekly.com
/Articles/2008/01/07/228771/manchester-airport-lifts-one-bag-rule-with-oracle.htm

It’s a bit of wishful thinking but imagine the ease in travelling if every airport harnesses the capabilities of Siebel Analytics and lifts the above mentioned rule.

This first post of mine in the ‘Analyzing Analytics’ series is an attempt at exposing the readers, be it a neo-techie or a seasoned expert, to the vast capabilities of the end product of Siebel Analytics: The Intelligence Dashboards and the concept of Dashboard centric business operation and management. So let’s discuss what these ‘Dashboards’ are all about.

Case 1: What are the sales for this quarter? What’s the comparison of Sales for this quarter with year ago sales? How many customer accounts are in the red? Such kinds of questions are very common in day to day business analysis. The answer to these and a plethora of similar process oriented questions helps in running a business efficiently.

Case 2:Organizations often have data fragmented across multiple sources like spreadsheets, department specific systems, etc. Often critical data is buried deep somewhere in some file or system in the organization.

Keeping the above two cases in mind, a unified view of the organization is the need of the hour. An Analytical dashboard is just what any organization would desire. It helps in giving a holistic view of business operations, most often using data that has been gathered into a warehouse and occasionally working from operational data. Pre requisites to a healthy dashboard are clean data, a good data model, sturdy data mappings and neat analytics.

Essentially, a dashboard is the final outcome of a Siebel Analytics operation. Based on the metaphor of the instrument panel in a car, the Siebel Analytics version of a dashboard provides a user, mostly business managers, with the input necessary to “drive” the business. Siebel Analytics Interactive Dashboard refers to a web-based technology page on which real time or warehoused information (Reports) is embedded from various sources used in the business.


As seen in the above example of Interactive dashboards, business data is displayed infused with an array of user friendly views like Pie Charts, drill-downs, navigations, pivot table summaries, bar graphs, meters and gauges . These dashboards are often role-driven and customizable. They can be modeled to show historical, current and predictive data which leads to measure – based decision making. The biggest advantage of an Interactive dashboard is that it helps in decision making being more fact based and less intuitive.

Dashboard centric management is definitely going to be the next best thing in business analysis. The idea of viewing and analyzing the organizational facts and figures with the help of superb layouts in front of your computer screen is absolutely fantastic. In short, Dashboards are all about infusing ‘Intelligence’ in the business of the organization and a wonderful apparatus for any organization to possess.

Read More about Analyzing Analytics