SAP Analytics CLOUD – SAP HANA Studio – How to optimize business applications performance by simplifying HANA view monitoring in one Digital Boardroom?

CONTEXT

To effectively monitor reporting solutions and underlying HANA view objects used in your IT project, SAP Analytics Cloud can help you to highlight information such as broken views, broken procedures, attributes details, dimensions details, dependent objects, input parameter details and more. SAP Analytics Cloud provides analytical capabilities, that help you quickly control the quality of your business application’s performance. Often, SAP analytics developers use SAP HANA Studio to manage the SAP HANA database, to create and manage user authorizations, and to create new or modify existing models of data in the SAP HANA database. With SAP HANA Studio you can check the status of your views by running complex queries to find a solution to your issue. If you want to avoid repeating the same queries several times in SAP HANA Studio, you can use SAP Analytics Cloud to retrieve all this data in a single digital boardroom. This will allow developers to improve decision-making by performing their tasks faster.

ISSUE

I am part of an analytics development team. My team must monitor our analytical solutions (called Solution A and Solution B in this blog) and is proactively trying to identify potential issues in the data flow, such as procedure runtimes, broken objects, parameter settings… Currently we monitor these issues by running SQL statements in SAP HANA Studio every day. Since we do this activity directly on the database, the monitoring can only be performed by a technical expert. Furthermore, completing such checks with SQL in HANA is not optimized: we need to run an SQL statement, display a table, and finally interpret the result. So, we decided to migrate the checks to SAP Analytics Cloud. Doing so also provides transparency to our business colleagues, ensuring collaboration and accountability across the entire team.

GOAL

Quickly identify and analyze data through HANA views instead of using complex queries in SAP HANA Studio

To facilitate the explanation of the different steps, I will focus on the example of broken objects.  For other use cases, the method is similar to the one laid out below (Data Validation => Data Extraction => Data Visualization).

Broken object means, that there is something incorrect with the HANA view or with underlying views or tables. It could be, for example, that a field is missing. If an object is broken, it affects activation of other objects.

APPROACH

• Data validation

Before consuming data from SAP HANA Studio, the first step is to create your own repository of the project in your environment. The creation of the calculation views is based on tables, table functions, and scalar functions specific to our solution’s content. Make sure, that you are creating your calculation view of type Cube because only then it is supported by SAP Analytics Cloud using  Live Connection. In this example, HANA view CL_CORP_SOLUTIONA_BROKEN_OBJECTS contains information about the count of broken views.

In the first projection, we are making use of a HANA system table called ACTIVE_OBJECT and applied a couple of filters as the package_ID (refine the search for views) and the object_status different to 0.

On HANA, any of the objects have a status: if it is “0”, that means the objects are fine  If an object has any status except “0,”  such as “1” or “2”, the object is broken.

Next, we calculate the counts of procedure, views, tables, and roles functions that are broken. This step occurs in the aggregation node.

“OBJECT_SUFFIX” field provides the service of the objects. So far, we are considering all views like “analyticview”, “calculationview” and “attributeview.” All broken objects are counted in this step.

We used a dummy table to prevent the view having no record when there are no broken objects. With this step, we will be able to see in the dashboard the count as 0. It is important to keep the key flag property to TRUE so that we always get data from this dummy table.

Finally, we combine all data with a union. In the last step we create an additional column called object type. This helps us in differentiating whether the object is of type view or procedure. It is just a simple calculated column. Leveraging data preview directly in HANA Studio, you can have a first preview of the result.

• Data extraction

In the next step, you can create the connection between SAP Analytics Cloud and SAP HANA Studio. In our case, we used a LIVE connection, that allows us to retrieve the real time status of objects from HANA.

I built a model that sits on a LIVE connection consuming data through SAP HANA Studio. The connection is created by selecting the SAP HANA system type and the connection that is required.

Then, to select the source data, I entered the name of the HANA view which contained the data of our broken objects. Data modeling is one of the most important steps to be consumable in a Story. Once the model has been created and saved, you can modify your preferences, selecting the desired dimensions and measures.

NOTE: For further help documentation on setting up Live Connections to SAP HANA on SAC, please see the link here.

• Data visualisation

Creating a Story

After creating the models, you can build an SAC Story to visualize the data. To display the main information and to have a quick current overview of the system health, the Solution A Overview tab allows to quickly identify the status of key performance indicators.

For our use case, we developed several additional page that help to analyze any issues deeper by exploring in detail.

For example, the Broken Objects page shows in detail, which objects are broken by Object Type. For the analysis of broken objects, depending on the scope of the project, it is preferable to add a filter according to the Object Type, so that you can determine which objects are broken.

The interaction between the different charts allows you to specify the content you want to display. For example, for the Procedure Monitoring page, by selecting the ‘Linked analysis’ option, we can filter by day or by procedure name.

Creating a Digital Dashboard

To create a dashboard that brings together the necessary information from several solutions, the idea is to create a Digital Boardroom combining all KPIs relevant for us in one application. In the Digital Boardroom functionality of SAC, there are two types of presentation available: agenda and dashboard. In our case we chose type Dashboard.

By clicking on Dashboard, you will open a modelizer for your digital boardroom. In the Library tab, you can import your story and start to structure how you want your dashboard to look like. You can either pick all the tabs from your story or add only those you want to your digital dashboard.

You can also add some filter or some jump from page to another page. In the presentation setting, you can adjust your preference, for the layout and configure your context menu. Once you are finished with your design, your digital boardroom is ready to be used.

CONCLUSION

What is the benefit?

• It is Cross platform that can be used by different devices
• Available to everyone (Business and technical user)
• You can access the dashboard on a mobile phone
• Using standard solution (no custom solution)
• With one digital boardroom we have access to all the different KPIs that are relevant for us
• Help technical users to quickly manage some incidents, resulting in faster response to our internal customers