Technology Blogs by Members
Explore a vibrant mix of technical expertise, industry insights, and tech buzz in member blogs covering SAP products, technology, and events. Get in the mix!
cancel
Showing results for 
Search instead for 
Did you mean: 
Former Member
5,562

Many organizations (prepare to) migrate their existing SAP BW landscape to SAP HANA. A migration to SAP HANA is always ‘non-disruptive’, meaning you can migrate to HANA as-is, unlocking many benefits instantly. However, simply migrating to HANA without optimizing your system would discard the full potential of the new database platform for BW. This article explains a 4-step approach to optimize your SAP BW landscape and benefit of all the innovation HANA has to offer.


Step 1: Instant gain: Performance!

First of all and most obviously, you will experience a significant performance improvement compared to traditional databases like Oracle or DB2. This performance gain has several aspects.

  • Front-end query performance

Thanks to SAP HANA’s in-memory and column-based character, reads on the database are much faster compared to non-HANA BW. This enables you to not only speed up your badly performing queries, but also to design queries on datasets or with calculations that could be executed before because they would time-out or run for ages before returning results.



  • Back end performance

Long runtimes in daily data loading and processing are often a big problem in SAP BW environments. Problems during nightly data loads are often the cause for delays in data availability, resulting in business users not being able to run their critical reports in time, or not at all. SAP BW on HANA offers great improvements in performance for BW back-end loading processes, significantly reducing the runtime of nightly data loads. When running on HANA, these runtimes can be reduced, and problems can be addressed earlier and faster, resulting in quicker data availability for reporting and far less business reporting downtime.

To ensure you benefit the most of performance improvements, it is important to migrate to the latest SAP BW release. New improvements and optimizations are added with every release. For example: the most recent, SAP BW (7.4 SP9), contains numerous new functionalities improving the SAP BW landscape and performance.

Performance statistics as collected by several of our clients after the HANA migration (before further optimizations):

Performanceresults

Step 2: Identify optimizations

Optimizing your BW on HANA system should be the first step after migrating. Some sweet spots:

  • Optimizing ABAP code for HANA

As SAP HANA is a column-based database, different rules apply when it comes to performance-optimizing your custom code. Performance-optimizing your ABAP code can make all the difference in loading performance after migration. There are several checks and programs SAP delivers that identify possible optimizations and give hints on what can be improved.



  • Redefine daily process chain schedules

The increased back-end performance HANA offers makes it possible to execute the daily loading process much faster. By redefining the process chain schedule and setup, you can put business-critical process chains to the very start of the schedule. This makes it possible for BW support teams to fix failed loading processes long before they negatively impact critical business reports.



  • Making transformations ‘HANA-Enabled’

SAP BW 7.4 SP8 on HANA introduced the ‘push down’ functionality for Transformations to the database, greatly increasing the loading performance of these transformations. The functionality in transformations that is supported to run on the database however is quite limited as of yet. Custom ABAP coding in in transformations for example are not supported yet, and not all formulas are supported yet, though more formulas are being added in every SP. If a transformation contains any non-supported functionality for pushdown to HANA, the check functionality in the transformation will point out what functionality is not supported.

Making slight adjustments to your data model in order to maximize the number of HANA-enabled transformations can make quite the difference in loading performance.

Pushdown

Step 3: Simplify your landscape/minimize redundancy

The best way to reduce throughput times for transformations and activations is to just get rid of them. When running on HANA, there is no performance benefit for InfoCubes anymore. In fact, making an InfoCube HANA-Optimized actually makes it very similar to a DSO from a table point of view.

In classical BW environments, lots of InfoCubes have been created with the sole purpose of performance, and are simply a 1:1 copy of the data in the DSO layer underneath it. By making a few minor adjustments to the DSO, and redirecting your MultiProvider (or Query) to the DSO instead of the InfoCube, you can decommission the InfoCube completely. This makes for 100% performance improvement for that step, simplifies the data flow, and also removes a place where errors can occur. This can be a huge benefit for data flows with logically partitioned InfoCubes or use of a Semantically Partitioned Objects.

Step 4: Use new functionality

SAP BW on HANA offers a variety of new possibilities that enable organizations to move on from the classical BW Data Warehouse to what SAP calls ‘the virtual datawarehouse’. With HANA’s processing power, data models will not only be simplified. There is also a range of possibilities to virtualize data layers in your BW data model. By not using persistence in every step of your BW Data model, development speeds up, flexibility is increased, and memory footprint is reduced.

With SAP BW on HANA, SAP has (finally) made modeling based on non-SAP data sources much easier compared to previous versions of the product. By using field-based modeling, Smart Data Access and the possibility to use Open ODS views, using BW to combine SAP with non-SAP data is now a very good option. The possibility to expose BW providers as HANA models also provides the possibility to use SQL on BW infoproviders, making BW more suited to be consumed by SQL-based front-ends.

SAP HANA supports many very advanced libraries for, for example, extensive statistical uses. Libraries like R, PAL and AFL can be consumed using a HANA Analytic Process in BW, and provide the result of those analytical processes back to BW.

For a full overview of new functionalities in the newest release of SAP BW, please see What's new in SAP BW 7.4 SP8

11 Comments
Labels in this area