Translate

Three things to know when migrating SAP BW on SAP HANA

Three things to know when migrating SAP BW on HANA

to ensure a smooth migration of the existing SAP BW System from Release 7.0 (including SEM, BPC and IP) and onwards the following three topics must be considered.
  1. shrink the data which will be moved to the row store of the SAP HANA database with the SAP BW housekeeping task lists
  2. shrink the data which will be moved to the column store of the SAP HANA database with the SAP-NLS Solution
  3. use the database migration option (DMO) as part of the SUM tool for the in-place migration including upgrade/migration/unicode* steps, see the Blog from the DMO Product Management.
    please note that all nZx technologies available for SUM natively only, so they are not available with the DMO option for SAP BW due to the difference between an Upgrade and a Migration.
DMO_3things.JPG


The on Premise use case – minimized production downtime with the combination BW-PCA and DMO

migration_all_options.JPG


The HEC use case – moving BW on Premise to the HANA enterprice cloud with the combination BW-PCA and DMO

migration_HEC.JPG
further Information can also bve found in the following Documents (each of them is an own blog itself)

 SAP First Guidance Collection – BW on HANA

Also the new SEM Add-On modeling on top of the NetWeaver BW System allows a much quicker creation of the needed stack.xml for the DMO procedure.

Database Migration Option for Software Update Manger (DMO)

  • DMO simplifies the procedure of a migration to SAP HANA by combining the upgrade and the migration
  • DMO is general available with SUM SP09
  • DMO may cover the Unicode Conversion (single code pages) with SUM SP10
  • DMO can be used for ABAP systems that are part of either SAP BW or SAP Business Suite
  • DMO includes a system update. This means that at least one SP has to be applied and a stack.xml is always required.

Purpose

This page will help you to prepare for an upgrade using the Database Migration Option (DMO) for the Software Update Manager (SUM),to prevent any issues that will prevent the successful completion of the update and  troubleshoot issues that occur during the Add-on Handling phases


Overview

Helpful notes, troubleshooting tips, known issues and phases of the DMO for SUM

Helpful Notes to prepare for the upgrade

This is a list of notes that will be helpful to review before starting the update, and during to help with troubleshooting issues
2257362- Database Migration Option (DMO) of SUM 1.0 SP17  
1968508 - Release Change & Single Code Page Conversion to Unicode with DMO
2345636 - SAP Solution Manager 7.2: Simplified Upgrade and Migration to SAP HANA  (NEW!)
1799545 - Using DMO of SUM for SAP BW systems
1875197 - Using DMO of SUM for SAP Business Suite systems
1968508 - Release Change & Single Code Page Conversion to Unicode with DMO
1724496 -Latest patches for R3-tools
1897665 -R3load: declustering support for tables in / namespace

Helpful Note for resolving problems during the upgrade


2054965 - R3load: TOC for logical table is incomplete in declustering mode after restart
1978565 - Upgrade with database migration option: MAIN_SWITCH/PARMVNT_XCNV; incorrect SQL statements
1959955 - SUM upgrade/migration to HANA fails during HDB_MIGCONFIG
1897665 - R3load: declustering support for tables in / namespace
2096202 - Row store/column store inconsistencies after system migration with DMO
1447267 - Error messages that can be ignored in RUN_RADCUCNT_NEW
1908075 - BW on SAP HANA: Table placement and landscape redistribution
1644396 - SMIGR: No data export for aggregate tables
1981718 - R3load: support for CDS views and some fixes
2236178 - Dump "DBSQL_TABLE_UNKNOWN", Table T439I doesn't exist in the Database after upgrade

Helpful Notes for troubleshooting

1666976 - uniqueChecker usage description 
1784377 - Checking pool tables and cluster tables
 

Troubleshooting tips

  1. Use the latest patch of SUM, most known errors are resolved with the latest SUM patch
  2. Use the latest kernel, kernel tools and DBSL, ensuring that they are patched in both <SUM_DIR>/abap/exe and <SUM_DIR>/abap/exe/2_nd
    Most known errors are resolved using the latest patches of the upgrade tools, kernel and kernel tools (R3load, R3szchk etc)
  3. Analyze the log files, the reason for the stop error will be written to the log file (see table below).

Performance optimization

SAP note 2153242 - Estimation of table sizes and downtime for SUM with Downtime-Optimized DMO
Please note, cluster tables may take longer to migrate in phase EU_CLONE_MIG_DT_RUN than standard tables.

DMO Phases in SUM and Error analysis

When?
Phase name
Description 
Directory
Logfiles
Uptime
EU_CLONE_MIG_UT_SIZES
 Determination of table sizes
 for uptime migration 
-
EUMIGRATEUTSIZES.LOG
 
EU_CLONE_MIG_DT_SIZES
 Determination of table sizes
for downtime migration 
-
EUMIGRATEDTSIZES.LOG

 
EU_CLONE_MIG_UT_CREATE
 Tables are created on target database
for uptime migration
migrate_ut_create
EUMIGRATEUTCREATE.LOG +
MIGRATE_UT_CREATE_RUN.LOG
 
EU_CLONE_MIG_DT_CREATE
Tables are created on target database
 for downtime migration 
migrate_dt_create
EUMIGRATEDTCREATE.LOG +
    MIGRATE_DT_CREATE_RUN.LOG
 
EU_CLONE_MIG_UT_PRP
Creates directories and control files for uptime
migration  (STR, TSK, CMD files for R3load)
migrate_ut
EUMIGRATEUTPRP.LOG

 
EU_CLONE_MIG_UT_RUN

Transport of tables into target database  
migrate_ut

EUMIGRATEUTRUN.LOG +
MIGRATE_UT_RUN.LOG
 
EU_CLONE_MIG_DT_PRP

Creates directories and control files for downtime
migration (STR, TSK, CMD files for R3load) 
migrate_dt
EUMIGRATEDTPRP.LOG

Downtime

EU_CLONE_MIG_DT_RUN

Transport of tables into the target database 
migrate_dt

EUMIGRATEDTRUN.LOG +
MIGRATE_DT_RUN.LOG

Known Issues

Kernel tools R3Load, R3ldctl, etc. should always be up to date at the latest patch stack available at the Service Market Place
Update both <SUM_DIR>/abap/exe and <SUM_DIR>/abap/exe/2_nd  folder to avoid problems.

Related Content


Related Documents


Related SAP Notes/KBAs

1895764 - R3load hangs if called with unknown option
2099073 - R3load writes an incomplete header in the TOC file of logical table after resta
1818067 - Merge of R3load, R3ldctl, R3ta packages in R3tools
1914260 - Abort R3load task if TOC file has no entry for the table
1854483 - R3load does not create primary keys in declust. by import
1905316 - R3load skips table with error message "no entry for table"
1981718 - R3load: support for CDS views and some fixes
1949365 - Various improvements in R3load and R3szchk
2097926 - R3load keep blanks option for R3load anonymous mode and other corrections
1859676 - R3load option '-sort_cluster' and action modifier 's'
1923294 - R3load adds extra spaces to strings after unicode conversion
2023495 - R3load reports source IDs and other corrections
1958957 - R3load cannot import some tables when performing Unicode conversion from Asian
1963554 - Improvements in R3load and R3ldctl
1931360 - R3load improvements for declustering of empty table clusters
2044380 - R3load dependency loop if calles with -o option and other corrections
1891426 - R3load on HANA to avoid unnecessory MERGEs during import
2095316 - R3load corrupts DDNTF table when changing the database
1897665 - R3load: declustering support for tables in / namespace
2017805 - R3ldctl: fix CDS views creation in upgrade export mode
1816978 - R3ta: incorrect split ranges to tables with RAW fields
1778564 - R3load, R3ldctl, R3ta Development Changelog
1824246 - R3ldctl duplicate index if TADIR definition is missing
1809726 - R3ta split rules may contain holes
1933687 - R3ldctl fails with assertion if called with decluster

Rebuilding SAP Indexes

When an indexed field in a table is changed repeatedly, the data is deleted from the index and re-inserted at a new block in the index. The new entries are always added to the right side of the index tree and therefor deletion is relatively to the left.

When the index elements are deleted and added repeatedly, the left side of the index tree tends to be sparse and the right side keeps growing. Depending on the size of index, there might be thousands of sparse blocks, which are being checked during execution.

Fragmented index results in increased usage of database space and more blocks being read into the buffer. This can be avoided by rebuilding the index.

One can measure index fragmentation using the report RSORATAD or using DB02 --> Detailed Analysis --> Enter Index --> Detailed Analysis --> Analyze Index --> Storage Quality.

If storage quality is less than 50% you may need to reorg the index. If you wish to run an analysis on all the indexes, run the report RSORAISQN.

Check SAP note 970538 for more details on the restriction with using this report. Do not run the report without reading the note.

You can also get an idea on the amount of fragmentation by comparing the size of the table and the index. If the size of the index is larger than that of the table, the index is heavily fragmented.

To rebuild index, you can use one the following methods:
1. brspace -f idrebuild -i <index_list>
2. Using DB02
DB02 --> Detailed Analysis (in the "Tables and Indexes" section) --> Object Name: <index> --> Detailed Analysis --> Alter Index --> Coalesce / Rebuild
3. Using RSORAISQ report
   Call SAP transaction SE38 and execute the report RSORAISQ.
   Input the tablespace name and the index name in Object Selection section.
   Check rebuild option in Function Code section.
   A report will be generated with details on the index.
   Click on the Index Rebuild button.
   You will be prompted to feed the name of the index and tablespace again, enter the details and execute.
4. Rebuild indexes offline using the report RSANAORA
5. Mass rebuild of indexes using the report RSORAISQN



If you find this useful, please consider sharing the post with your friends and colleagues on facebook,twitter, google+, linkedin (or any other medium you like!)

Would you like to read more on indexes? The linked post discusses guidelines to creating an Oracle index based on table size.

Monitoring Table and Index Fragmentation (Oracle)

Fragmentation of tables and indexes may reduce performance, depending on the way data is accessed. Fragmentation also leads to greater overall storage space usage. This problem can be eliminated by reorganizing the particular object. However you should bear in mind that this process is very expensive and that the system is not available during a reorganization. It is often not advisable to immediately start a reorganization to eliminate fragmentation. For more information on when to perform a reorganization, see Reorganization.

Table Fragmentation

Table fragmentation will result in longer query times when a full table scan is performed. Since data is not as evenly packed in the data blocks, many blocks may have to be read during a scan to satisfy the query. These blocks may be distributed on various extents. In this case, Oracle must issue recursive calls to locate the address of the next extent in the table to scan.
Recent studies have shown that table fragmentation has hardly any effect on the performance of the database system. This is mainly because full table scans are somewhat rare in an SAP system since data is accessed using an index. Reorganizing table data is generally not as beneficial to performance as previously thought. For more information on how to perform a reorganization, see Reorganizing Tables with BR*Tools

Index Fragmentation

Index fragmentation may bring a higher penalty to application performance. When accessing data through an index and an index range scan (common in SAP systems), Oracle must read each block in the specified range to retrieve the indexed values. If the index is highly fragmented, Oracle may have to search many more blocks, and possibly levels, to get this information. To eliminate index fragmentation, you must rebuild the index.
In both cases – table fragmentation and index fragmentation –always make sure that the soft and hard limits for the number of extents (MAXEXTENTS parameter) are not reached. If this happens, you must intervene.

See also:

SAP HANA Cloud Integration for Data Services – an ASUG Webcast

Ben Hofmans, HANA Cloud Integration product owner, more specifically for Data Services provided this ASUG webcast.   As background you can  read Dick Hirsch’s blog and comments What if HANA Cloud Integration was already productive and nobody knew (or cared)


HANA Cloud Integration is an on demand version of Data Services.  The usual disclaimer applies – things discussed below are subject to change.  Please note these are my notes and I am not a subject matter expert in this area.

1fig.png
Figure 1: Source

Figure 1 shows that there different kinds of clouds

Integration is important in all cloud types

On the left of Figure 1 is the Managed Cloud where SAP provides hosted version of software  - HANA Enterprise Cloud is an example – you get a HANA box – it is not in your network, you don’t acquire hardware or maintain.  It is by SAP in the managed cloud.

In the middle blue box – LOB applications – SuccessFactors, ByDesign, Sales & Operations Planning, Travel on Demand – you don’t have access to HANA database but the UI’s to connect to cloud applications.

Third area is Marketplace – launched at SAPPHIRE – you buy SAP applications – mainly Amazon and other partner clouds provided later

There are many clouds. Integration is an important service, says Ben.

2fig.png
Figure 2: Source: SAP

LOB (“line of business) users is on the left and the right is IT with their requirements – for integration between on premise and in cloud

LOB users want a single source of truth and master data synchronization – one master system for employees, customers .

LOB needs more real-time business process integration

IT's main concern is data security.

3fig.png
Figure 3: Source: SAP

On premise in the cloud and on premise systems and third party solutions

SAP HANA Cloud integration includes process integration and data integration

NetWeaver PI & Data Serivces for data integration - the Cloud brings these two together, with real time process integration.

It has pre-packaged integration flows from SAP system on premise, with a  place to share content

SAP HANA Cloud Integration (Data Services)


The same development team who works on premise works on the cloud offering.

This is Data Services in SAP HANA Cloud, which is a new solution, new architecture, for multi-tenant cloud environment

4fig.png
Figure 4: Source: SAP

SOP is Sales & Operations Planning and since last December is live and GA

It can move data from on premise (ERP APO) and load to HANA hosted in cloud – where SOP planning is built.

SOP  enables organizations to align supply & demand in real-time

5fig.png
Figure 5: Source: SAP

The SuccessFactors application is used to manage employee performance

For business execution use BizX

You can get data from on premise or the other way around, as shown in Figure 5.

6fig.png
Figure 6: Source: SAP

You can access SAP HANA box and SAP HANA marketplace, with your HANA database, data services

You can use this to extract data from on-premise (Business Suite, or data bases) and load to Hana database

7fig.png
Figure 7: Source: SAP

Figure 7 shows Data Services on Demand, available today.  It provides a role based user interface

Data Services extractors that work for on premise also work for Cloud, said Ben.

8fig.png
Figure 8: Source: SAP

Figure 8 shows that there is a lightweight installation inside – where to install, which user to install

The agent packages and sends through https to the cloud

You do not need to open firewall  - the request is from on premise to the cloud

9fig.png
Figure 9: Source: SAP

Ben said you do not need a hard core development.  It offers an HTML5 interface as shown in Figure 9.

You can drag and drop to model data flows from on premise to the cloud or the other way

During this webcast, Ben provided a demo of integration.ondemand.com

10fig.png
Figure 10: Source: SAP

Later this year they plan to offer a license not tied to an SAP cloud application

Limited ETL – only focused on what cloud applications needed (this does not include unstructured)

11fig.png
Figure 11: Source: SAP

Ben said Figure 11 was presented at SAPPHIRE.

SAP Lumira is the new BI on demand solution – need to load to cloud – Hana cloud will play a role for delta loads

Subset of Question & Answer

Q:  Is "Data Services On Demand" only applicable when loading data onto the SAP cloud, or could be used to load data onto an on premise SAP system ?
  A:  For the moment, the Cloud needs to be one end of it. It can load into the on premise system.
________________________________________________________________

Q:  Does this provide integration for SAP BW and SF Workforce Analytics?
A:  Workforce Analytics is in pilot. Not GA yet.
A:  Currently it is not available yet for SAP BW
________________________________________________________________

Q:  SAP Cloud for Travel - we are very close to purchasing Travel On Demand - should we ask our SAP sales person about whether HCI would be the way to integrate with on-premise ERP?  I assume it would be included in ToD pricing?
A:  Definitely worth enquiring.
________________________________________________________________

Q:  How is performance of the data loading ensured?
A:  There are 3 parts that influence performance: (1) extraction, (2) data transfer to the cloud and (3) inserts in the HANA target database.

The onPremise agent is based on the SAP Data Services engine, so has a long history of optimizing extraction from various sources through native protocols (RFC, business content extractors and ABAP generation, as well as native database access for non-SAP systems).
For the data transfer to the cloud, the onPremise agent will compress the data before sending it to the cloud. This is also the part of the overall loading time where bandwidth between the customer’s network and the cloud plays a major role.

Once the data is received by the HCI server component, a dedicated HANA loader will take care of high performance inserts/updates in the HANA database in the cloud.
________________________________________________________________

Q:  In which country is the cloud data stored?
A:  SAP has Data Centers across the world.
________________________________________________________________

Q:  I assume Hana Cloud Integration can schedule tasks, can one task call another task? Resulting in the triggering of Tasks, based on other task completion.

A: Yes, HCI can schedule tasks.
Scheduling is currently time-based only, it is not possible to set dependencies between tasks, or launch one task from another task.

Labels

sap hana hana database aws s4 hana hana db s4hana conversion steps sap hana azure bw4hana hana migration s4hana migration sap cloud migration steps sap hana migration steps sap hana migration to azure s4hana sap fiori fiori performance fiori erp s4 hana fiori sap fiori app sap fiori client sap fiori launchpad sap s4 hana fiori cisco ecc AI SAP AI abap dumps hana sap S/4HANA S/4HANA Conversion best sap ui5 & fiori training ccmsidb configuration database fiori tutorial on webide free sap ui5 & fiori training s/4 hana sap dumps sap fiori tutorial sap ui5 sap ui5 & fiori sap ui5 & fiori tutorial sara ui5cn 1514967 2367245 - Troubleshooting performance issues with SAP BPA Amazon free tier for SAP AWS setup Experience CALL_FUNCTION_NOT_FOUND CCMS Configuration and Use Create New Data Class in SAP (Oracle) Critical top SAP Abap dumps DHCP Clients Not Receiving IP Addresses Download Stack.xml HAN-DB HAN-DB-ENG Handling Spool & TemSe High CPU Usage Due to Excessive Process Switching How To How to Start and Stop SAP Hana Tenant Database How to change SAP Hana Sql Output results are limited to 5000 Records How to perform SAP Dual Stack Split - Netweaver Inactive Objects in SAP Intercompany transactions in SAP AP / AR : Cross Company Code Transaction Interface Flapping Due to Duplex Mismatch KBA LOAD_PROGRAM_LOST MSSQL shrinking transaction log file Migrating to SAP hana database NAT Overload Causing Internet Access Failure Note 500235 - Network Diagnosis with NIPING OSPF Adjacency Not Forming PRINCE2 Foundation Sample Questions Preparing for S/4HANA Conversion and the MUST know items Push to Download Basket S/4HANA Central note S/4HANA Migration Cockpit S/4JANA SAP BI Support Data Load Errors and Solutions SAP BI/BW Landscape SAP BPA SAP Basis SAP Basis Automation SAP Business Objects SAP CPS SAP Certification SAP FI Certification SAP FI Certification Sample Questions SAP HANA Admin - Cockpit SAP HANA DB Engines SAP HANA Database SAP HANA terminate session connection disconnect cancel kill hang stuck SAP Hana DB restore SAP Hana Numeric Error Codes SAP IS-Retail SAP Landscape SAP Language installation SAP MM and Purchase Order Tables SAP Maintenance Planner SAP Note 500235 SAP R/3 Glossary SAP Readiness Check SAP S/4HANA 1709 Installation Files SAP S/4HANA 2023 SAP S/4HANA 2023 Installation SAP S/4HANA 2023 running SAP S/4HANA Installation SAP Scheduling SAP Solman 7.2 CHARM: SAP Support Package Stack Strategy SAP Support package SAP Upgrade SAP support stack upgrade SP stacks STORAGE_PARAMETERS_WRONG_SET SUSE/SLES/Kernel versions Setup of S/4hana 2023 TSV_TNEW_PAGE_ALLOC_FAILED TSV_TNEW_PAGE_ALLOC_FAILED error Transaction ID Unable to download an SAP Note Unix/Linux Command That Are Helpful For SAP Basis Upgrading SAP Kernel Without Downtime Upgrading windows server 2008 to windows server 2019 What is OSS Notes? SAP SNOTE Tutorial accounting agile ale idoc ale/edi archive FI documents audit auditing auditor aws aws cloud basic type bluefield approach ccms charm copilot datavard dbacockpit download sap note download snote edi idoc electronic data interchange enable sap archiving objects erpprep ffid firefighter fraud functional hana admin how to apply sap security note https://www.erpprep.com/ idoc install install sap fiori installation interfaces intermediate document internal control license key linux version materials management messsage niping test order type port prince2 agile prince2 agile practitioner purchasing s/4hana upgrade s4 hana sap abap dumps sap abbreviations sap activate certification sap activate project manager sap authorization sap aws sap background job sap brownfield sap ccms sap ccms configuration sap erp sap error sap grc sap greenfield sap internet demo system sap kernel sap license sap maintenance certificate sap material management sap meaning sap mm sap mm consultant sap monthly security note sap netweaver sap network diagnostic sap niping sap note sap oss sap patch day sap performance sap performance issue sap purchase order sap s/4hana sap sales and distribution sap sap otc sap sd sap sd certification training sap sd course sap sd jobs sap sd module sap sd online training sap sd training sap sd tutorial sap sd tutorial for beginners sap security sap security note sap snote sap snote tutorial sap solution manager sap spooler sap sql segregation of duties separation of duties sles slicense sm37 smc snote snote in sap system sod conflict solution manager solution maneger stop start hana database suse linux techie trex two step upgrade required update sap kernel waterfall