Tweet
Blog --> SAP HANA Data ProvisioningLogin To Create Post

SAP HANA Data Provisioning

- By Prakash Joshi

There are many ways to propagate data in SAP HANA. There are three fundamental options:

  • SAP Landscape Transformation (SLT), also called trigger-based replication
  • SAP Data Services, also called ETL-based replication
  • Sybase Replication Server (RS), also called log-based replication

Any of these can be used independently, or all three can be used together. Depending on the technical characteristics of the source system and on the desired data provisioning strategy, you can implement the appropriate mechanism for each of your sources of data.

In addition to these three methods, you can also leverage existing extractors; that is, Business Content data sources originally developed for use with SAP NetWeaver Business Warehouse (BW). This is called Direct Extractor Connection (DXC) and only covers extraction from data sources, so it’s more of a complement to the others, rather than a replacement for any of them. DXC generally applies more to BW on SAP HANA than to SAP HANA as a standalone implementation for analytics.

Each of the three main data provisioning mechanisms is fundamentally different from the others.

SLT/trigger-based replication is the primary data-provisioning mechanism for SAP source systems, for example, SAP ERP. When considering SLT replication, below are some key facts:

  • SLT requires the installation of an SLT Replication Server.
  • Both data and metadata from tables can be replicated to SAP HANA with this technology.
  • The SLT Replication Server records any changes to a set of designated tables inside the source system and makes them available for a pull into the SAP HANA database.
  • The SLT Replication Server intercepts the changes to the tables in the application layer when transactions are executed in SAP ERP by detecting the triggers that are sent to the database to update tables, making this solution database independent on the SAP ERP side.
  • The replication of data to SAP HANA can be set up according to a schedule or as a continuous, near real-time feed.
  • You can filter data within a table; for example, when replicating the accounting tables, you can specify which company codes you want the data for and which fields should come across to SAP HANA.
  • EE You can also perform limited transformations during the data export from the source system using the SLT Replication Server. More complex transformations are technically possible, but pay special attention to performance with this functionality to avoid impact on the replication performance, especially for continuous data feeds that you want to use for operational reporting.
  • The SLT Replication Server is an ABAP-based technology that leverages several proven existing system landscape optimization technologies from SAP, which are now updated to work for SAP HANA.
  • The SLT Replication Server can also be used for non-SAP sources that meet some basic database criteria that enable the SLT Replication Server to capture the changes in the source system.

SAP Data Services/ETL-based replication uses the SAP Data Services technology that is already in place in many implementations. When considering SAP Data Services replication, below are some key facts:

  • This requires the installation of the SAP Data Services software component.
  • The replication can leverage existing extractors, function modules, and programs in the source system.
  • Both data and metadata can be replicated to SAP HANA with this technology.
  • Replication of data typically runs on a more traditional batch job schedule, for example, every few hours or once per day.
  • SAP Data Services is often used to integrate non-SAP source systems into the reporting landscape.
  • Highly complex transformations and data-cleansing capabilities are possible.
  • SAP Data Services is a proven, highly stable technology for which many companies already have the necessary tools and skill sets in house.

Sybase Replication Server/log-based replication is a database-specific technology that is available only for specific database systems, namely the Sybase and IBM DB2 for LUW (Linux, UNIX, and Windows) systems. When considering Sybase Replication Server (RS) replication, below are some key facts:

  • RS starts from the database level, using database log tables to identify any changes to tables in the source system.
  • This is a high-performing, real-time replication technology that bypasses the application layer.
  • Because the application layer is bypassed, there are no opportunities for filtering or transforming data; the mapping is one-to-one and at the table level.
  • The data is copied to SAP HANA exactly as it’s entered into database tables in the source system. The database commits are essentially replayed on the SAP HANA database.

High level Architecture Comparison: -

  SLT SAP Data Services Sybase Replication Server
Installation Requirement SLT server must be installed either on a separate machine or on an SAP ERP application server. SLT server is based on SAP NetWeaver 7.02, kernel release 7.20 EXT. SAP Data Services 4.2 SP2 must be installed. This is a regular SAP Data Services implementation; no special components need be installed on SAP Data Services to work with SAP HANA. Sybase Replication Server (RS) must be installed on the SAP HANA server. Sybase Replication Agent (RA) must be installed on the source system server.
Number of Source Systems Multiple source systems to any number of SAP HANA systems. Multiple source systems to any number of SAP HANA systems. One source system per SAP HANA instance.

Data Replication Capabilities Comparison: -

  SLT SAP Data Services Sybase Replication Server
Data Movement Real-time and scheduled replication. Scheduled replication. Real-time replication.
Data Replication Approach Replication set up on the table level. Replication according to SAP Data Services configuration. Replication by Logical Unit of Work.

Presence of Load Balancing Load balancing (parallelization). Load balancing (parallelization). No load balancing.
Data Transformation Limited data transformations and filtering are possible. Complete ETL toolset for data transformations. No transformation of data.
Supported Tables Transparent, pool, and cluster tables can be replicated. Transparent, pool, and cluster tables can be replicated. Transparent tables can be replicated. No support for pool and cluster tables, but consult SAP Notes.
Support of Compressed Tables Compressed tables are supported. Compressed tables are supported. Tables with compressed values are supported for DB2 9.1, 9.5, and 9.7. Row-compressed tables are only supported for DB2 9.7.

Source System Compatibility Aspects Comparison: -

  SLT SAP Data Services Sybase Replication Server
SAP Systems SAP ERP systems from 4.6C, other ABAP-based systems from Basis release 4.6C. For SAP systems, this is an RFC connection from SLT to the source system. SAP ERP systems from 4.6C; other ABAP-based systems from Basis release 4.6C. SAP ERP 6.0.
Non-SAP Systems Any system on SAP-supported database (note: DB2 for iSeries and Sybase ASE as of SAP HANA 1.0 SPS04 only). For non-SAP systems this is over a DB connection. Any. Feasible in RS, but currently not in scope for SAP HANA.
Unicode Unicode or non- Unicode source systems. Unicode or non- Unicode source systems. Unicode source systems only.
Database Source Support Source system can be on any SAP-supported database. Source system can be on any SAP-supported database. IBM DB2 for LUW only.

Administration and Configuration Aspects Comparison: -

  SLT SAP Data Services Sybase Replication Server
Administration Administration in SAP HANA Studio. Administration via regular SAP Data Services Management Console, or through SAP Solution Manager. Administration in SAP HANA Studio using the Load Controller component.
Configuration Some configuration in SLT server, especially with first setup of new replication scenarios. Configuration in SAP Data Services. Configuration in RS.
Transformations Transformations are configured and processed on SLT server. Transformations are configured and processed in SAP Data Services. No transformations possible.
Sorry,No Comment To Display:     
Name:
Email-Id:
Write Comment:
Share This Blog:
 
 
Copyright © 2008-2016. All rights reserved. BDI Logo