Blogs

SAP HANA views organised in layers

17 augustus 2018

Getting structure in the new HANA environment is not as easy as it first seemed. It all started easy, just one model to combine some ADSO generated views and use it in Analysis for Office. But after these scenario’s some new requirements popped-up: the simple views needed an update and became more and more complex. Some logic from the ADSO was removed and put into HANA views, meaning that the previously used generated views had to be remapped. That got us thinking… We needed a more efficient way to avoid this in the future.

Every layer connects to the layer underneath with the use of a contract layer. This takes care of the mapping between two layers. A view in one layer does not care what happens in the other layer (and should NOT be used in another layer), the only thing they need to respect is “the contract” in between. A model can change to whatever is necessary as long as it respects the contract. In doing so everybody using the contract knows that anything being build on top of it will work after updates or changes down the line.

Analytic layer

Objective: Reporting layer, exposed views for the reporting tools.
Concept:

  • Business-reporting logic
  • Data Category: Cube
  • Small calculations
  • Replacement of BEx
  • Generic name, understandable for business users

Output layer

Objective: Contract between Analytic and Model.
Concept:

  • Contract layer
  • No logic, calculations, joins…
  • Data category: [blank]
  • Generic name, understandable for business users
  • Proper naming of fields
  • Translations

Model layer

Objective: Development layer: create / combine / apply company logic.
Concept:

  • Data category: [blank]
  • Building blocks
  • Apply logic and calculations
  • Create complete BI solutions

 

Input layer

Objective: Decouple the data from the system.
Concept:

  • Contract layer
  • No logic, calculations, joins…
  • Data category: [blank]
  • Generic name, explaining what data is inside
  • Remapping from SAP to Hadoop possible without breaking logic above

Raw data layer

Objective: Data from every source possible.
Concept:

  • Generated HANA views from SAP
  • SQL tables
  • Hadoop tables
  • Smart Data Access
  • Blogs

    Open Hub Destination with BEx Query as an InfoProvider

    Introduction

    A SAP BW system is mostly known to consolidate data to give reporting or analytics capabilities to your business. However, there could be occasions where you want to transfer the data. One of the options to do this is to make use of Open Hub Destination. This was in the past a part of the InfoSpoke, but separated since SAP BW 7.X. The Open Hub Destination allows you to distribute data in database tables or flat files in a fully integrated and controlled SAP BW data flow:

    Read more >
  • Blogs

    Getting started with SAP HANA SDI in less than a day

    (The Twitter case) Part 2

    I was recently asked by a client’s marketing department to figure out how to bring Twitter data to their reporting environment. They are currently using a BW-on-HANA, so I immediately thought of HANA Smart Data Integration (SDI) as a possible candidate. In this blog I’ll cover the basics for the setup and show you that it’s actually quite easy to get up-and-running with SDI. In fact: I was pulling data from Twitter less than 3 hours after starting this SDI adventure 😊.

    Read more >
  • Blogs

    Getting started with SAP HANA SDI in less than a day

    (The Twitter case) Part 1

    I was recently asked by a client’s marketing department to figure out how to bring Twitter data to their reporting environment. They are currently using a BW-on-HANA, so I immediately thought of HANA Smart Data Integration (SDI) as a possible candidate. In this blog I’ll cover the basics for the setup and show you that it’s actually quite easy to get up-and-running with SDI. In fact: I was pulling data from Twitter less than 3 hours after starting this SDI adventure 😊.

    Read more >