Watch the following video to learn more about the multisystem usage scenarios supported by SLT.
Note
Note
- The new CDC mechanism is no longer limited to four targets using the same set of triggers. It is now truly 1:N.
- The new CDC mechanism is used by default in case a new configuration with an S/4HANA 2020 source system is created. There may be reasons to switch back to the old CDC mechanism for some special scenarios (for example replicating INDX-like tables). See documentation of report IUUC_REPL_SWITCH_REC_MCHNSM for more details.
If you are using a DB source, then the restrictions and improvements do not apply. In that case, please refer to SAP Note 1768805- SAP Landscape Transformation Replication Server (SLT): Non ABAP-based Sources.
With DMIS 2011 SP15 & DMIS 2018, it is possible to create multiple 1:1 configurations in addition to a 1:N configuration or another 1:1 configuration. This functionality was down ported to DMIS 2011 SP13 and SP14 with the 2603957 correction note and for S/4HANA 1709 FPS00 with the 2609517 correction note.
Find more information in note 2648409under the Corrections section.
There is a possibility to create multiple 1:1 configurations in addition to a 1:N configuration or other 1:1 configuration.
Allow to have more than four consumers for one table for the trade-off of worse performance on the source table due to additional triggers. It is therefore highly recommended to use this option only when there is no other choice and you understand the potential consequences for your scenarios running in the source system. Overall, it is not recommended to use this setup for a productive landscape which has a high change rate on the source tables or batch jobs performing bulk operations on these tables. Enabling multiple 1:1 configurations is possible by setting parameter 'MULTI_1_1' in table IUUC_RT_PARAMS to 'X'.
Installation Options
Before beginning the installation, it is important to understand the various system landscape options:
- For an ABAP-based SAP Sources system, the SAP SLT can be installed on the source system.
- SAP SLT can be installed on a dedicated standalone ABAP-based system.
- For an ABAP-based target system, it is possible to install the SAP SLT Server on the target system.
Note
Installing the SAP LT Replication Server as a separate system is recommended for the following reasons:
- It allows a clear separation of concerns - the SAP Landscape Transformation Replication Server system will only replicate data and all system resources are dedicated to this.
- It allows a different set of users to be the single users to have access to the system (security aspects).
- It allows independent planning of upgrade cycles or other maintenance activities on the SAP Landscape Transformation Replication Server system. For example, this will avoid downtime in the Central Finance system if any maintenance activities need to be done in the SAP LT Replication Server system.
- It supports the independent sizing of servers and allows you to manage performance efficiently.
- It helps to keep an overview of the system landscape if multiple source systems are connected to one or more target systems.
The figure compares installation options, and outlines their advantages and disadvantages. A dedicated SLT server is highly recommended, especially for the productive landscape.
There is a fourth option, which is not explicitly outlined in the figure. This option is to run the SLT server on an SAP ABAP-based system that is neither the source nor the target system in your landscape, such as your SAP Solution Manager.
Product Versions and Supported Use Cases
Watch the following video to learn more about the SAP Landscape Transformation Replication Server for SAP S/4HANA 1.0.
SAP LT Replication Server as part of SAP S/4HANA 1610 – 1909
The following figure shows the scenarios supported on SAP S/4HANA 1909.
The following figure shows the scenarios supported on SAP S/4HANA 1610 - 1809.
For SAP S/4HANA, it was possible to use certain scenarios as part of the SAP S/4HANA system. In general, using this option to run SAP Landscape Transformation Replication Server is not recommended.
Embedded Installation Option
| Product Name | Support Basis Version |
|---|---|
SAP LT Replication Server for SAP S/4HANA 1.0 | S4 Core > = 105 (SAP S/4HANA 2020) |
SAP LT Replication Server as part of SAP S/4HANA 1610-1909 | S4 Core > = 101 (SAP S/4HANA 1610) < = 104 (S/4HANA 1909) |
SAP LT Replication Server can be used as part of an SAP S/4HANA system. The product is already part of the SAP S/4HANA installation. This is known as the embedded option. It is not possible to update SAP LT Replication Server independently of SAP S/4HANA.
Standalone Installation Option
| Product Name | Support Basis Version | AddOn Name | SAP Note |
|---|---|---|---|
SAP LT Replication Server for SAP S/4HANA 1.0 | S4 Foundation > = 105 | DMIS 2020 | 2954022 |
SAP LT Replication Server 3.0 | Netweaver = 7.52 S4 Foundation > = 102 < = 104 | DMIS 2018 | 2669326 |
SAP LT Replication Server 2.0 | When used as SAP LT Replication Server System:
When used as source/target system:
| DMIS 2011 | 1577441 |
The installation of the SAP Server as a separate system or as part of the source or target system via the DMIS AddOn for an SAP NetWeaver / SAP S/4HANA foundation-based system is known as standalone option because SAP LT Replication Server can be updated independently of other software components.
| ABAP / Non ABAP source system | SAP LT Replication Server* | SAP HANA System |
|---|---|---|
Installation
| Installation
| Installation
|
* The minimum SAP NetWeaver release is SAP NetWeaver 7.52.
See Notes:
- 3130804– Release Information SLT - DMIS 2018 SP07
- 2669326 – Release strategy and Maintenance Information for the ABAP AddOn DMIS 2018_1_752
- 2675613 – SLT (DMIS 2010 / DMIS 2011 / DMIS 2018 / DMIS 2020 /S/4HANA) Installation, Corrections, and Release Information
- 2577774 – Version compatibility for source and target systems – SLT
- 1605140 – SAP Landscape Transformation Replication Server (SLT)
As seen in the installation options, the SLT server can be installed on any SAP ABAP-based system in your landscape. A separate installation is recommended for productive scenarios. The previous figure shows the requirements for the source system, the SLT server, and the target SAP HANA database. Note that the ABAP source systems should have the same SP of DMIS 2018 installed as the SLT server itself. The prerequisites for other scenarios, such as an SAP ABAP-based target system, SAP Business Warehouse, or central finance, can be found in their respective units. The latest version and dependency information can be found in the Product Availability Matrix (PAM) or in the relevant release notes.
Note
SLT Background Work Processes
SLT works with background work processes and uses different types of jobs. To understand the concept of the load and replication procedures in more detail, you need to become familiar with the major type of jobs for SAP LT Replication Server, having a look at the Application Operation Guide.
Note
Note
Starting from SLT 2.0 SP15, in order to improve overall performance and reduce resource consumption by SLT, the SLT Master Job /1LT/IUC_REPL_MSTR has been removed. Logic which happened in the Master Job has been moved to the following jobs:
/1LT/IUC_REP_CNTR_ <MTID>
Master Controller Job
/1LT/IUC_HOUSKPNG
Housekeeping Job
/1LT/IUC_HEALTH_C
Health Check Job
Overall, tables will switch from "scheduled" to the following state much faster, especially if many configurations are present in the system.
Health Check Framework
There is a set of health checks that the system runs in order to alarm about unusual behavior of SAP LT Replication Server and potential issues or inconsistencies. Health Checks are either system checks or table checks.
Please check the existing health check jobs on the Application Operation Guide.
Average SLT Throughput
The data transfer rates outlined in the figure can be assumed, as long as the following measures have been taken:
- Default settings are used for reading types.
- No extra tuning or individual configuration is in place.
- The ideal table characteristics are maintained according to the SLT sizing guide's table categorization principle.
In the initial load, one background work process on the SLT server can transfer 10 million records per hour from a transparent table into a target system. For a pooled or cluster source table, use factor 0.6 = 6 Mio records per hour and work process. Once the records in the source table have been transferred completely, the replication phase begins. In data replication, one work process on the SLT server can transfer 1,000 records per second from the source system to the target system. For pooled or cluster tables, use factor 0.6.
The relationship between the number of data transfer jobs in the configuration settings and the number of available work processes for SAP Application Server for ABAP on the SLT server can be summarized as follows:
- Data transfer and data transformation processing on the SLT server system is accomplished in background work processes of the underlying SAP Application Server for ABAP.
- This requires sufficient CPU and memory resources (for the work process content).
- The number of dialog work processes in the source system corresponds 1:1 with the number of data transfer jobs in the SAP Landscape Transformation Replication Server system.
As well as the work processes allocated by the data transfer jobs, you need to provide additional available work processes for the following:
- Controller and monitoring jobs
- Migration objects definition
- Access plan calculation
- Configuration changes
With the exception of replication via the ODP framework and replication with replication logging active, no database persistence is used for data loading or replication.
SLT sizing is an exercise to determine how many work processes are required to do the following:
- Perform the initial loading of data into the target system within an acceptable timeline.
- Accomplish change capturing and transfer of data changes within the expected latency time to the target system.
- Add enough additional work processes to allow other required SLT jobs to run.
The overall runtime (RT) for the initial load of a table (set of tables) is determined in the following way:
- NR stands for the total number of loading-relevant records in all replication-relevant tables.
- The runtime depends on the number of utilized parallel jobs NJ, which are required to achieve the expected overall throughput.
![Two points under Replication case: Number of Jobs (NJ) = recorded data changes (T) [records/second] / 1000 (for Transparent tables) and Number of Jobs (NJ) = recorded data changes (T) [records/second] / 600 (for Cluster, Pool tables)](https://learning.sap.com/service/media/topic/c0729882-e7a6-4338-8364-9598f86c13c2/SLT100_19_en-US_media/SLT100_19_en-US_images/Unit_2_-_Lesson_1_Landscape_and_Sizing_010.png "Two points under Replication case: Number of Jobs (NJ) = recorded data changes (T) [records/second] / 1000 (for Transparent tables) and Number of Jobs (NJ) = recorded data changes (T) [records/second] / 600 (for Cluster, Pool tables)")
The number of data transfer processes that is required to accomplish a target data transfer latency of 1 second is determined in the following way:
T is equal to the total number of replication-relevant data changes in all replication-relevant tables per second.
You can perform a detailed sizing of your SLT server and the other affected systems using the following input parameters:
- Number of configurations
- Number of tables per configuration
- Details about each table, including the following:
Table type [transparent/cluster]
Number of records [rowcount]
Size of single record ( < > 1500 bytes/record)
Numbers of columns (S: < 150, M: 151....250, L: > 250)
Expected change rate [changes per hour]
Complex data transformations required [y/n?]
Max. tolerable initial load time [hours]
Max. tolerable replication latency [sec]
Watch the following video to learn more about the installation options.
Note
Note