Discover how to choose the right data-warehousing platform for corporate reporting and planning. This comparison of the capabilities of SAP BW, SAP HANA, and SAP BW on HANA helps elucidate the best use and benefits of each platform.
Key Concept
SAP BW Analytic Manager (formerly known as OLAP processor) orders SQL requests in a proper sequence, assembles query results, and pushes down query execution to the SAP HANA server. SAP BW Analytic Manager compiles the execution plan, opposed to the SAP BW system that stores the metadata and is used for the transport and the lifecycle of the data model. BEx Queries are compiled by the Analytic Manager into an optimized execution plan, meaning that most parts are executed in SAP HANA, and only a small part in the SAP BW application server.
When I first heard about SAP HANA and its enhanced performance, I was very interested in discovering more about the product and learning about the benefits of its in-memory capabilities. It seemed as if SAP HANA would deliver on the most critical of the issues BI projects need to solve: improving query run time and data-load performance. It also seemed that data loads may not be needed at all as all data is sitting in memory. I have many years of implementation experience with SAP BW. This, coupled with my recent experience implementing SAP HANA, made me decide to put together a summary of my findings to provide an overview of the benefits (if any) of implementing the new SAP in-memory technologies, and to discuss which tools can be used for what purpose.
I was especially interested in learning about how the logic delivered in SAP BW 7.x transformations with ABAP routines are implemented in SAP HANA, and if it really performs thousands of times faster as promoted by SAP. On the other hand, I wanted to find out for myself if the new SAP HANA platform has the same flexibility and powerful reporting and planning capabilities as SAP BW 7.x has.
The answer to my first question is, yes. You can still keep the old ABAP logic in SAP BW transformations when migrating to SAP BW on SAP HANA. However, without optimizations to the system, the performance of these routines is not substantially improved as the routines are executed on the application server in the same manner as they are without SAP HANA.
As far as gaining more flexibility in designing reporting and planning solutions, a lot more flexibility is gained with the new SAP HANA platform, but not as much as is gained with SAP BW 7.4 on SAP HANA. As it stands today, SAP HANA by itself won’t be able to replace all SAP BW’s functionality and reporting or planning capabilities, but I believe that SAP BW 7.4 on SAP HANA provides an additional dimension for corporate reporting and planning.
Step by step, I review some of the new features and capabilities that make SAP BW 7.4 on SAP HANA unique and, in many ways superior to, other data-warehousing platforms. At the same time I discuss SAP BW on HANA by comparing it against the performance of SAP BW on any other database and the SAP HANA platform itself.
Although there are a number of possible deployment scenarios, I assume that a generic high-level technical design (as outlined in Figure 1) is applicable to most SAP BW on SAP HANA deployments, with SAP BusinessObjects used as a front-end tool.
Figure 1
Sample technical design of SAP BW on HANA deployments
Using Master Data, Authorizations, and SAP BW Analytic Manager
Most reporting scenarios use master data navigational attributes, texts, hierarchies, authorizations, and OLAP processor functionality for data drill downs, analysis, and data crunching. Technical users and developers should be able to configure and operate master data objects, set up authorizations, and design complex queries. Table 1 is a brief summary of the capabilities of each platform in this respect. This comparison illustrates how these platforms compare with each other in terms of handling master data, setting up authorizations, and designing/executing complex queries.
| SAP BW on any database |
SAP HANA |
SAP BW on HANA |
|
Uses built-in functionality for InfoObjects attributes, texts, and hierarchies. BEx Query Designer uses standard functionality for complex query definitions, such as formulas, variables, exception aggregation, conversions, virtual characteristics, and key figures. Having these capabilities in place makes it easier for developers to design complex queries faster.
In addition, SAP BW analysis authorizations allow access to configure analyzed data by data slices according to business rules.
|
Requires custom development for each specific scenario. This may lead to substantially longer solution delivery. Counts and exception-aggregations logic can be implemented via complex SQL statements. This can lead to substantially longer development or maintenance efforts.
Pros: SQL requests are optimally designed for SAP HANA execution. In this scenario the best query performance is achieved using the SAP HANA calculation engine.
Cons: This solution requires a lot of custom development for simple reports, which may result in a longer lead time for solution delivery.
SAP HANA authorizations can be leveraged as opposed to with SAP BW, which uses SAP BW analysis authorizations.
|
In addition to the standard SAP BW functionality, the capability to expose BW models to SAP HANA and do native development on them is gained, and the models can then be directly consumed by BI front-end tools, such as SAP Lumira, BusinessObjects Explorer, and SQL.
SAP BW Analytic Manager orders SQL requests in a proper sequence, assembles query results, and pushes down query execution to the SAP HANA server. |
Table 1
A comparison of master data and OLAP processing capabilities by platform
Data-Load Scheduling and Monitoring Capabilities
Any database or data warehouse loads data from sources and, in some cases, sends processed data to destinations. It is nearly impossible to set up complex data-loading scenarios without adequate tools that support data-load monitoring and scheduling. Table 2 shows the data-loading tools that are currently available by platform and the pros and cons of each.
| SAP BW |
SAP HANA |
SAP BW on SAP HANA |
|
Open-hub functionality: SAP BW has built-in open-hub functionality that allows you to distribute data from a BI system to non-SAP data marts, analytical applications, and other applications. It ensures controlled distribution across multiple systems.
You can read more about SAP’s open-hub functionality in detail in the SAP Help portal: SAP Help.
Process-chain functionality: SAP BW has a powerful process scheduling and control capability called a process chain. A process chain is a sequence of processes that are scheduled to wait in the background for an event. Some of these processes trigger a separate event that can, in turn, start other processes.
You can read more about SAP’s process chain functionality in detail in the SAP Help portal: SAP Help.
Monitoring and statistical capabilities: SAP BW has a series of process loading, statistical, and monitoring capabilities, including technical content for statistics, system tables, and process chain run-time analysis. The sequence of process chain executions can be monitored in transaction code ST13. |
If you are using SAP HANA as a transactional ERP system, you may want to use it for operational reporting as well, as you then do not need to load data into another system. However, this scenario relies fully on transactional data and accurate business content (e.g., sets of views and procedures that translate database tables into something meaningful for business users). SAP HANA business content started showing up in certain functional areas and has to be investigated case by case.
At the same time when using SAP HANA as a data-warehouse platform you can leverage the built-in job scheduling tool in the SAP HANA XS Administration Tool. You can read more about this tool here: SAP Help.
Admittedly, at this time (March 2015), the scheduling tool in SAP HANA XS Administrator is not as advanced as the process chain tool in BW (Figure 2). |
In addition to the standard SAP BW extract, transfer, and load (ETL) scheduling and monitoring functionality, the new SAP HANA Analysis Process monitoring and scheduling capabilities can be fully integrated in BW process chains and monitored from there (see the sample process chain in Figure 3).
For more details about the SAP HANA Analysis Process go here: SAP Help.
|
Table 2
A comparison of job scheduling and monitoring capabilities by platform
Figure 2
The job scheduling tool in SAP HANA XS Administrator
Figure 3
Sample built-in process-chain monitoring capability of SAP BW
The Role of Business Content
One of the main reasons companies have chosen to go with SAP BW for corporate reporting has been SAP BW’s comprehensive Business Content capability, which substantially reduced the time for deploying BI solutions. Business Content in SAP BW (commonly referred to as SAP BI Content) and extractors are a set of predefined objects (e.g., InfoCubes, Relational DataStore Objects [DSOs], data sources, and queries) that help implement BI reporting solutions faster.
How important is Business Content in SAP BI? Obviously, it may be missing for certain scenarios or functional areas depending on the platform. The question is, is its absence going to be an impediment to a successful implementation? I would say it depends on the reporting area. In rare cases, where there are relatively simple requirements based on a few database tables, you may be able to design reporting solutions without using BI Content. However, in the Finance, Logistics, and SAP Human Capital Management (SAP HCM) functional areas, for example, the number of database tables is huge, and the table relationships and business interpretations are rather complex. In these cases designing reporting solutions from scratch may be very costly and time-consuming.
A set of financial and logistics extractors delivered by SAP cannot be easily replicated by custom applications simply by linking database tables in SAP ERP Central Component (ECC). Table 3 is a brief outline of SAP BI Content availability by platform. Availability or absence of BI Content may dramatically impact the speed of BI solution delivery.
SAP BW on any database
|
SAP HANA |
SAP BW on SAP HANA |
|
SAP has been working to improve Business Content in SAP BI for BW during the last 15 years by adding new business meaning to the complex technical configuration of ERP. Business Content in SAP BI offers users numerous advantages, most of which target corporate strategic reporting, including the following:
- Reporting data model templates
- Translation of technical tables and fields into business language
- Delta capabilities of standard extractors processing huge datasets
- Master data enriched with attributes, text, and hierarchies
Traditional SAP BW has BI Content that usually serves as a design starting point in most BI projects.
|
What about BI Content for SAP HANA? What if you have ECC on SAP HANA and want to have real-time operational reporting based directly on ECC data?
SAP has introduced its SAP HANA Live rapid-deployment solution that enables real-time reporting on operational data from SAP Business Suite applications. This rapid-deployment solution provides pre-built reporting content and pre-configured content for easier and faster analysis of operational data on the SAP HANA platform. You can read more about it here: SAP HANA rapid deployment.
For a number of industries, SAP HANA-based repositories (data models) are available with SAP HANA content. For instance, for SAP Retail there is an SAP Customer Activity Repository (CAR) that provides a common foundation and a multi-channel transaction data model for consuming applications. You can read more details about it here: SAP Help.
SAP HANA content for CAR is a collection of views on the data stored in the SAP HANA database. Follow this link to learn more: SAP Help.
|
In the last few years SAP has invested substantial effort in delivering SAP HANA-optimized BI content. SAP HANA-optimized BI Content combines the enterprise data-warehousing capabilities of SAP BW on SAP HANA with explorative and interactive real-time analytics using the SAP HANA in-memory database. The new data model follows the recommendations for the layered scalable architecture (LSA++), which is optimized for SAP BW on SAP HANA.
For more details about SAP BI Content for SAP HANA go to the SAP Help portal: SAP Help.
|
Table 3
A comparison of BI Content availability by platform
Non-Cumulative Key-Figure-Based Models
When working with inventory reporting and stock analysis, you have to use proper modeling capabilities in order to efficiently address reporting performance and deal with fluctuating stock movements.
In this example, imagine that you have 1 million product articles, 100 plant/storage location combinations, and approximately 5 million stock movements per day. To be able to design reports on daily stock balances, you have two options:
- Store stock balances for each article code, plant, storage location, and day
- Store opening balances per article and all movements per article. Recalculate balances per article for every day at query run time.
If you choose option 1 (store daily balances), you end up with an issue related to processing an unnecessary volume of data. This is because you would have to store 1 million balances per day for each plant and storage location. If you have 100 plant/storage location combinations, you end up with more than 100 million records per day. This volume is vast even by the standards and capabilities of SAP HANA.
In the second scenario (store opening balances and movements) you would have to load opening balances by article once and store daily movements (e.g., 5 million records per day). This is more practical from a data-volume point of view. However, as balances are recalculated at report run time, you may end up with an issue with query performance. This second option is known as a non-cumulative model, as you have to use non-cumulative key figures and special non-cumulative InfoCube functionality in SAP BW. Table 4 shows how this approach is addressed in each platform.
| SAP BW on any database |
SAP HANA |
SAP BW on SAP HANA |
|
SAP BW has built in non-cumulative key figures and data flows that are usually used in inventory management and HCM headcount reporting. This functionality has certain pros and cons, briefly listed as follows.
Advantages of non-cumulative modeling:
- The fact table is kept smaller (by factors)
- The history remains (e.g., all inventory movements)
- Has the ability to report by both levels at a point in time and movements for a certain period
Disadvantages of non-cumulative modeling:
- The query run time can be affected by the calculation of the non-cumulative model
- Deleting transaction data for material that is no longer current is not possible because deletion cannot be restricted by time
Here is a link to the SAP Help page that describes this non-cumulative concept in detail:
SAP Help.
|
The details about the virtual data model recommended by SAP for inventory management in SAP HANA can be found at this link: SAP Help.
There are a number of views available for inventory management of stock quantity in SAP HANA based on ERP. Follow this link to learn more details about this solution: SAP Help.
It is important to know that, in order to stay within reasonable data volumes, ERP inventory movements usually get archived on a regular basis. In the case of retailers or large distribution companies, inventory movements may be kept only for a few months, especially in the case of Stock Transfer Orders (STOs).
While SAP HANA models can be used for operational inventory checks and current reports, SAP BW can be used as an EDW platform for inventory analysis on historical data.
|
In BW 7.4 Analytic Manager (Support Package 8) query execution related to handling of inventory key figures is pushed down to the SAP HANA engine. This means that inventory-level calculations in the non-cumulative scenarios are performed by the SAP HANA engine instead of the application server, which leads to faster query executions.
|
Table 4
A comparison of non-cumulative modeling capabilities
Planning Tools Requiring SAP BW
Both the SAP Business Planning and Consolidation (BPC), version for NetWeaver and the Integrated Planning (IP) planning tools require SAP BW. SAP BW on SAP HANA is closely integrated with BPC and IP, which makes it easier to harness the performance power of SAP HANA in complex planning scenarios. When using the Planning Application Kit (PAK), most IP models have a much better planning performance in BW on SAP HANA due to the processing being pushed down from the application layer into the SAP HANA database and calculation engine.
Here is some additional information about SAP planning applications (e.g., IP and BPC) in relation to SAP HANA: https://www.saphana.com/docs/DOC-2144.
Manage Existing Investments in ABAP and BW Models
When migrating from SAP BW to SAP BW on SAP HANA you need to know how much configuration can be directly reused on the new platform. In many cases there is extensive ABAP logic in extractors, transformations, BEx variable exits, and so forth. Most of the logic can be kept in BW on SAP HANA. However, for those elements that are the most critical for performance, I recommend that a complete redesign be done, taking into account the specifics and benefits of the SAP HANA platform.
Whenever possible, you should consider replacing ABAP routines in transformations with SQL logic in order to benefit from processing being pushed down from the application layer into the SAP HANA database and calculation engine.
Final Thoughts
Every existing SAP EDW platform has its own application area: SAP BW can work without SAP HANA on relatively small data volumes, SAP HANA itself can serve as a powerful in-memory platform for developing new custom applications, and SAP BW on SAP HANA combines calculation power of in-memory data processing together with the enhanced reporting, EDW, and planning capabilities of SAP BW. There are a number of new features and capabilities in BW 7.4 on SAP HANA, enabling a whole range of new reporting and planning scenarios and simplifying architecture (you can read more about them in my whitepaper: “A Few Cool Features in BW 7.4 on HANA that Make a Difference”).
In summary, here is a list of the capabilities that I believe make BW on SAP HANA a powerful EDW platform that will have a big impact on corporate architecture:
- Built-in master data modeling capabilities
- BW Analysis Authorizations
- BW Analytic Manager with BEx Query Designer
- Open hubs for connecting to external data outputs
- Process chains and monitoring tools for data load scheduling and monitoring
- Comprehensive Business Content for easy data extraction from SAP ECC
- Non-cumulative models for inventory and headcount reporting
- Composite providers and open ODS views for mixed reporting scenarios (SAP HANA and SAP BW on SAP HANA)
- Planning tools (IP and BPC)
Sergei Peleshuk
Sergei Peleshuk has more than 15 years of experience implementing BI technologies for global clients in retail, distribution, fast-moving consumer goods (FMCG), oil, and gas industries. He has helped clients to design robust BI reporting and planning capabilities, leading them through all project phases: from analysis of requirements to building BI roadmaps, technical architecture, and efficient BI teams. Sergei is an expert in modern BI tools and technologies available on the market, including SAP Business Warehouse (SAP BW), SAP HANA, BusinessObjects, and SAP Lumira. Sergei maintains a business intelligence portal at www.biportal.org.
Sergei will be presenting at the upcoming SAPinsider HANA 2017 conference, June 14-16, 2017, in Amsterdam. For information on the event, click
here.
You may contact the author at peleshuk@biportal.org.
If you have comments about this article or publication, or would like to submit an article idea, please contact the editor.
