Skip to Content
Author's profile photo Frank Schuler
Frank Schuler
March 15, 2012 2 minute read

Market research data analysis with SAP HANA

With a background in marked research data analysis we immediately identified this high data volume activity as a potential use case for SAP HANA. The HANA Developer Desktop evaluation then gave us the opportunity to implement our scenario.

Access to the HANA Developer Desktop evaluation is provided via CloudShareCloudShare.jpg

We use the Fullscreen RDP acces to the HANA Developer Desktop evaluation because it is the fastest option. However if your network does not allow RDP connections there is a browser based access as well.

RDP.jpg

Exclamation.jpg Please make sure to remeber the locale you chose when adding your user to your HANA Developer Desktop evaluation system because it will have to match any subsequent package you create. The SAP HANA Studio is quite peculiar in that it would consider English (United States) and English (United Kindom) two different languates in this respect.

We then created the tables that implement our data model which I will describe in more detail later

Tables.jpg

Exclamation.jpgPlease assure that all your table and column names are upper case otherwise you will get error messages that either your table or the respective columns do not exist.

and filled them with data via the ftp interface into the HANA Developer Desktop evaluation system

FileZilla.jpg

Exclamation.jpg Please make sure to set the permissions of the directory you create for your data files on the ftp server as well as the permissions of the data files themselves to -rw-rw-rw-. Otherwise you would not be able to save your data files into on the ftp server nor to read them into your HANA Developer Desktop evaluation system

The market research data we analyse has three dimensions (into which we loaded these instances) resulting in a data set of 3 million records

  • outlet (198)
  • period (108)
  • product (814)

DEMO_DATA1.png

Each data item has two measures

  • value
  • volume

and one calculated measure

  • average price

Measures.png

As a result it is possible to browse those 3 million records in fractions of a second in SAP BusinessObjects Explorer

Result.png

Many thanks to Alistair McCallum for leveraging his 40 years of extensive data and database experience to help with this piece of work. He currently holds the position of QA manager at Comprise Limited responsible for delivering quality products.

Assigned Tags

      Be the first to leave a comment
      You must be Logged on to comment or reply to a post.