You can create a data flow from one or more data sources. Data flows are a way to produce a curated data source that you can use to easily and efficiently create meaningful data visualizations. But have you ever wondered exactly what is happening when you execute your Oracle Analytics data flow? How can you see the exact log of actions triggered by the dataflow you just executed?
Most of us rely only on the system confirmation that an Oracle Analytics data-flow ran successfully along with the creation of a refreshed data set. But what if it fails? What if, the user is interested in knowing what exactly happened in the underlying database source, and what was retrieved in the Oracle Analytics engine?
A data-flow fires a set of 'logical SQL' commands to the Oracle Analytics server, just like any other data visualization queries do, and hence generates various logs behind the scenes that can be visited to track what exactly happened when it was run. But how can you find the right log? Here is a simple pragmatic trick to allow users a peek behind the scenes and do this is quick and simple.
Subscribe to the Oracle Analytics Advantage blog and get the latest posts sent to your inbox
In our example scenario, we need to track the logs of a data-flow we just built. The logs exist in the Oracle Analytics servers, but we don't know how to easily find them. A simple trick to help identify which log is the right one is to mark our data-flow with a searchable marker. That can be done by adding a new column with a unique name in the data-flow. In this screenshot below, we are adding a column named 'TIP33' for example:
Note that this addition to the data-flow will have no adverse effect on the resulting data. Let's just save the dataflow and re-run it now
Next, we open the Oracle Analytics console and navigate to Session and Query Cache.
Now let's see what happened.
Here we can see the server output of any activity within the Oracle Analytics server. To find the most recent queries pertaining to our Data-flow run let's just search the page for "TIP33."
This highlights entries in the log that pertain to our Data Flow. By looking up the logs, physical SQLs, numbers of rows returned, and potentially error messages, Users can now see what executed when the dataflow was run. You can now rapidly fine-tune your data flow designs to optimize them!
If you found this interesting, check out what's new in Oracle Analytics. Transform anything you screenshot or draw into an interactive data visualization. Turn any image into insights with digitized map layers and get greater control over the user experience. Also, see how Oracle Analytics for Cloud HCM gives HR cross-functional alignment with Finance along with self-service analytics capabilities.