SAP HANA Modeling in SAP HANA Support Package 9: Using the Rank Node in Graphical Calculation Views

  • by Ned Falk, Senior Education Consultant, SAP
  • April 6, 2015
Learn about the nuances of the SAP HANA Support Package 9’s rank node in graphical calculation views. Learn how to limit the amount of data to be sorted to just the important records, not the noise generated by your Big Data environment.
Learning Objectives

Reading this article you will:

  • Learn how to describe the purpose of the ranking node feature in SAP HANA graphical calculation views
  • Understand the basic deployment options for the rank node
  • Understand and visualize the effects of all the advanced options in the rank node
Key Concept

With the release of SAP HANA Support Package 9, the graphical calculation view information model has a few added features. One of the most useful is the new ranking option. Although SAP HANA offers fast data access, when a large quantity of rows is returned to the client tools, the total speed of the report can still be quite slow. Using the power of SAP HANA to rank and then return only the best (or worst) records gives users a more targeted result, faster.

Of all the new modeling features of SAP HANA Support Package 9 (SP9), the one I think is the most important is the new ranking node part of the graphical calculation view. Why is it great? Well, the answer goes to the core of a well-designed business intelligence (BI) system. The main purpose of a well-designed BI system is to give analysts the data they need to make decisions as fast and accurately as possible, while limiting noise. In this context, noise means extraneous data that does not help analysts make good decisions. The ranking node feature goes a long way towards meeting this goal by eliminating the records that are not important from the resulting analysis.

A very sophisticated, yet more complex way to meet this goal (e.g., giving analysts the needed data) is ABC Analysis, where an algorithm pre-processes the data and then physically assigns each value a code. Let’s use the following parameters to illustrate this scenario. In this example, there are 50 million sales records summarized by customer, and sorted top down by revenue. ABC analysis would assign customers that make up the top 20 percent of total revenue a value A. The next group is the value Bs, and the remaining customers make up the value C group.

With this ABC technique, you’re typically analyzing the data maybe once a week or once a month, then determining the rank as the ABC code, and persisting it as a field in the master data. For example, you would store the customer’s ABC code in the Business Partner Master tables along with the customer’s state, city, and ZIP Code. Filters in the initial query limit the data to normally just the A rank. Contrasting this method of eliminating the noise with the ranking option discussed in this article, the rank as a number (e.g., #1,#2, #3, and so on) is not persisted at all but rather is calculated in the run-time view. It is then used to limit the dataset to the top N or bottom N value, or, as in my example, the top three.

I address how to do this type of ABC analysis in SAP BW on HANA in more detail in a future BI Expert article. The ability to do this with native SAP HANA’s Predictive Analysis Libraries and the SAP HANA Application Function Modeler (graphical) also will be addressed in a subsequent article. Contrasting these data-mining tools with the subject of this article ranking yields advantages for both approaches.

Ned Falk

Ned Falk is a senior education consultant at SAP. In prior positions, he implemented many ERP solutions, including SAP R/3. While at SAP, he initially focused on logistics. Now he focuses on SAP HANA, SAP BW (formerly SAP NetWeaver BW), SAP CRM, and the integration of SAP BW and SAP BusinessObjects tools. You can meet him in person when he teaches SAP HANA, SAP BW, or SAP CRM classes from the Atlanta SAP office, or in a virtual training class over the web. If you need an SAP education plan for SAP HANA, SAP BW, BusinessObjects, or SAP CRM, you may contact Ned via email.

See more by this author


No comments have been submitted on this article. 

Please log in to post a comment.

To learn more about subscription access to premium content, click here.