Not long ago we released version 2 of our OBIEE performance and usage trend monitoring solution. In the meantime several interested parties send us a request to get the solution package and we installed it ourselves in our clients' OBIEE environments.
Naturally we'd like to share with you how OBIEE managers, architects, developers and administrators rate the reports provided:
"Really great! Thanks a lot for this free service."
OBIEE dashboard prompts can be configured to provide filters with operator BETWEEN. Image below shows an example on a day column.
After applying a low and a high value the query returns all rows with dates between low and high value including the edges, e.g.:
DAY BETWEEN low AND high
There are two additional options to utilise the between operator in OBIEE dashboard prompts without changing the operator in the prompt itself. You can just insert the low or the high value to get query behaviour as follows:
DAY >= low DAY <= high
Inserting just the low respectively left value turns the between operator into a greater than or equal operator. Applying the high/right value means to run a query with less than or equal operator. Of course, you can allow the user to select the operator via a dashboard prompt operator select list. That depends on your dashboard requirement. Be aware that using greater or less than operators in a selection may result in decrease of query performance.
As already OBIEE Performance Monitoring v1.1 also version 2.0 of our Oracle BI monitoring solution is deployed completely free of charge! Find the details for obtaining the configuration files at the end of this article.
OBIEE performance and usage trend monitoring is solely based on Oracle BI Usage Tracking. You can obtain everything described below for free (please refer to the end of the post).
OBIEE monitoring dashboard prompt now provides several filter possibilities to be able to search for detailed data. Thus you can for example analyse the origin of concrete performance issues or you can have a look to certain areas of interest.
Fields provided for filtering the monitoring reports are: Year: year of query start date Quarter: quarter of query start date Month: month of query start date Subject Area: the subject area of a query Dashboard Catalog Area: root catalog path of dashboard location, e.g. /shared/OBIEE_UK/ Analysis Catalog Area: root catalog path of analysis location, e.g. /users/gerd/ Dashboard Name: the name of a dashboard Analysis Name: the name of an analysis User Id: user login Day: a between filter for day level searching
Not all filters make sense for each analysis. But just don't use them in such a case. Catalog Area filters are very helpful when restricting queries to certain catalog storage locations. Keep in mind that filtering on Dashboard Catalog Area only returns queries initiated from opening a dashboard or dashboard page while Analysis Catalog Area calculates on queries released from a dashboard or from the OBIEE catalog. Ad hoc reports not saved in the catalog are not longer considered then.
Usage per Queries
Usage trend per queries shows number of queries within a time period (month, quarter or year). Charts illustrate overall query number, query numbers by dashboard, catalog and ad hoc usage and percentage share of these categories.
Category figures are mutually exclusive. They are calculated as follows: Dashboard Query #: number of queries initiated from a dashboard Analysis Query #: number of queries initiated from the catalog and not from a dashboard Ad hoc Query #: number of queries initiated neither from the catalog nor from a dashboard
Usage per Users
Trend per users equals ot usage trend per queries. Figures are based on distinct active users. Percentage share of categories doesn't apply here, because calculation isn't mutually exclusive for user numbers. A user can use Oracle BI in each category, therefore sum of the three categories is most likely greater than User #.
Period offers Start Hour. When combining start hour with day filter prompt (e.g. filter on today) you can see how many users are active intraday.
Usage per Top 10 Analyses
Top 10 analyses by number of queries including runtime average and average of data rows selected (due to privacy we cannot show names in the table chart).
Usage per Top 10 Dashboards
Top 10 dashboards by number of queries including runtime average and
average of data rows selected (due to privacy we cannot show names in the
Release Notes v2.0
The reports are integrated into two dashboards separated by performance and usage trend. Oracle BI repository and reports are based on OBIEE 184.108.40.206.1. For installation or downgrading support just contact us.
Please feel free to send us your specific analysis requests for Oracle BI report performance and usage trend monitoring. We will try to integrate them in the next release.
Obtain OBIEE Performance and Usage Trend Monitoring
As already mentioned we provide all relevant configuration files completely for free. Just contact us or send an e-mail to firstname.lastname@example.org. Please provide your company contact data. We don't use the data for marketing purposes. We just want to know who is using our OBIEE monitoring solution.
Blog article OBIEE Change Analysis Subject Area mentions by the way that you should use Column Formula editor for replacing field content in Oracle BI Answers. Main reason for this approach is to keep appearance of all views where field reference is used. You just want to change the field to show any other content having no effect on analysis' design. If you use the Delete button and add new columns again, then you must edit all views where field should be inserted. In most cases you will addtionally lose some nice settings in your views.
Let's take a simple selection of fields:
You possibly assume that fields shown link directly to the content delivered by OBIEE. That's not the case. Each field section just represents a specific column. It's a reference which is applied to the views in results tab. You can substitute the source of this reference. Subsitution doesn't affect view references. Therefore you don't lose any view settings when changing the source. Press Edit formula in order to change e.g. Query # figure:
Section Column Formula shows the source of the reference. Use subject area navigation on the left hand side or directly type in the new source, e.g. Runtime Avg observing the correct syntax, of course: