Dataset reload minimum frequency currently hour.
Designed a dataset through classic BI Analysis. Classic BI analysis populating from database. User looking when underline database table refreshed by ETL after that dataset also reload. As we have dataset reload API. Is this any option automating dataset reload through API.
It is always helpful to provide important details, such as the actual product you are using (OAC vs. OAS(versionX), how your query is configured (RPD, LSA dataset, etc.)
The (preview) REST API for dataset reloads is documented here:
This is for OAC DV, and there are documented restrictions.
For Classic Analyses via RPD, there are different strategies.
Here are a few blogs that may interest you. Using your favorite search engine may find more.
Oracle employees can also collaborate with their peers via internal channels.
They may state OBIEE, but would apply to OAS, as well
Other comments, welcomed.
Additionally, as an alternative to the reload data API for datasets, have you tried scheduling using the DV GUI?
Look at Scheduled Datasets Reload in Oracle Analytics:
Hi EzequielC-Oracle - Yeah that option we tried, but unfortunately lowest frequency is 1 hour. So, after ETL execute customer has to wait maximum 1 hour to refresh dataset. Due to that we are looking for a solution after underline database table of report refresh dataset reload also execute.
Hi @SteveF-Oracle - Thanks for sharing details. All documents are very informative. We are using OAS6.4 and designed dataset from multiple classic BI reports. Those classic BI analysis design from database which is refresh by ETL. After completion of ETL execution looking a solution to call dataset reload automatically without wait dataset reload schedule time.
Aren't you using the wrong tool for the job?
OBIEE first, and OAS now, has never been designed to be a database, to store data, and act as data source for analysis. The tool was meant as gateway to connect various sources together, model them to give them a business meaning, generate dynamic queries, and let business users ignore technical designs and features of a database and just perform analysis.
DV introduced datasets (in part because of a desperate need to be Gartner's compliant, in part because of self-service is often based on sources that aren't databases, Excel files for example), doesn't mean that you must use them. Because let's be honest: OAC/OAS has not a single chance to beat a database in term of performance or feature of data query, storage, optimization.
If your DV workbooks are built copying some classic BI analysis, why aren't your workbooks just based on subject areas?
If you do, caching should still be a thing (but I sadly wouldn't be surprised if DV just break the quite powerful server side caching available in classic) and could help. And here you should fall back on some of the links posted above by Steve on BI Server caching. (But here again: BI Server caching was often a quick-win for poor performing analysis because of design issues.).
Trying to use DV as a database and refreshing datasets after the ETL executed is, IMHO, a real misunderstanding on what the role of the product is, where it fits in a the whole data analysis architecture, and how it is built. Something that, sadly, the marketing message for the past years communicated in a poor way.
And if you really need to refresh a dataset triggered by an event (ETL completed), the whole DV is kind of built as a UI calling APIs in the backend. They just aren't documented, but you could easily look at the URLs called by the buttons/links you can click in the UI and script the same thing.
Thanks for clarifying that you you are using Oracle Analytics Server, connecting to a database via an RPD physical connection, and you are not using cached datasets (LSA, uploaded files).
You should be able to follow the cache purge techniques, after ETL run that updates your data in the database.
These techniques are well documented from a variety of sources with methods for triggering those purges after ETL update.
Hi
Please take a look at the recent KM doc that was published on programmatically clearing cache in OAC. Maybe you could schedule this script to be executed after every ETL refresh so oac cache is purged after ETL run : How To Programmatically Clear Cache in Oracle Analytics Cloud(OAC) using JDBC (Doc ID 2956513.1)
Thanks
Gayathri
