You have three key areas where you can boost BW performance on the back-end: the collection and extraction layer, the transformation layer, and the staging and storage layer. The author provides speed-up tips in all three of these areas.
Performance tuning of the BW back-end — the data model design and data flow across the model from the source system to the data targets — can produce real benefits. The goal of performance tuning on the back-end is to eliminate redundant processes and to reduce the volume of data to be processed during data upload from source systems to data targets.
You have three levels in the BW back-end to consider:
- The collection and extraction layer
- The transformation layer
- The staging and storage layer
At my company, we have learned a number of tips for improving BW performance at these levels, and 22 of them follow. For reference, Figure 1 provides a quick overview of how data flows from the source system to the BEx browser.
Collection and Extraction Layer
Data collection is performed on the source system. Prior to any data-upload activities, the extraction mechanisms and data source have to be configured on the relevant source system. The three types of data sources are SAP R/3 source systems, non-SAP source systems (such as an Oracle database), and flat-file source systems (such as a Microsoft Excel file). When extracting data from SAP R/3 source systems, you may use either Business Content extractors (which you can enhance) or generic extractors.

Figure 1
Data flow from source systems to the BEx browser
Business Content Extractors
Generic Extractors
Flat-File Uploads
Transformation Layer
Staging and Storage Layer
A few other aspects might affect back-end BW performance — your technical infrastructure and BW system hardware, for example. During the early phases of the BW implementation, make sure the following hardware sizing considerations are made: the number of current and future users, volume of data to be converted, and growth rate of the data to be transferred to BW from the source systems. If you have already implemented BW and found hardware constraints affecting the performance, consider upgrading your hardware.
Maintaining good performance on the BW back-end is an ongoing process. We recommend, therefore, that you take advantage of the tools that SAP provides to monitor BW performance. For example, activate and use the BW Technical Content statistical InfoCubes. They provide functionalities to monitor system performance at all levels (data upload, reporting, hardware utilization, and so on). Refer to Business Content documentation in BW Help for detailed information on statistical InfoCubes and their usage.
Nurullah Battal
Nurullah Battal has several years' experience in enterprise business solutions, with an emphasis on MIS and DSS systems based on SAP solutions. He is experienced in all phases of design and implementation of the end-to-end analytical applications on data-warehousing solutions, especially SAP SEM components on BW. In addition, Nurullah has considerable experience with SAP BW, SEM BPS, SEM CPM, SEM BCS, and BW/CRM integration. He also has experience with tuning the overall performance of very large systems. He has worked on implementation projects in North America and Europe. Nurullah is an accomplished BW/SEM instructor. Nurullah is a senior SAP BI consultant working for Turquoise Consulting, which specializes in SAP Business Intelligence solutions. He played a major role in developing Turquoise' SEM and BW implementation methodologies.
You may contact the author at nurullah.battal@turquoiseconsulting.com.
If you have comments about this article or publication, or would like to submit an article idea, please contact the editor.