DB6CONV 640 v47
DB6CONV 640 v47
DB6CONV 640 v47
1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Document History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4 Naming Conventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
8 Appendix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
9 Disclaimer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
This documentation describes how you can convert a single table or multiple tables using report DB6CONV for
SAP systems on IBM Db2 for Linux, UNIX, and Windows.
Table conversion here means that you move one or more tables within the same tablespace or to another
tablespace as fast and efficient as possible using Db2 means without explicitly changing the table structure.
Note
This documentation only applies to DB6CONV Version 6.40 or higher. For information about the changes
applied to this document, see the Document History [page 5].
You always find the most current version of the DB6CONV report attached to SAP Note 1513862 .
● Single table conversions are executed in one process, mass conversions (conversion of multiple tables) are
executed in one or more parallel processes.
● You can define conversions using either target tablespaces or target data classes.
● Support of data compression
You can set the compression attribute for target tables. The compression rate is optimized by means of
sampling.
● Support of index compression
For target tables, you can set the compression attribute separately for data and indexes.
● As of Db2 10.1, DB6CONV supports:
○ Adaptive compression for table data
○ Hidden columns
○ Insert time clustering (ITC) tables
● As of Db2 10.5 FP5SAP2, DB6CONV supports column-organized tables.
● Row-organized tables are online, that is, accessible for INSERT, UPDATE, or DELETE operations during the
conversion. Column-organized tables are read-only, that is, not accessible for INSERT, UPDATE, or DELETE
operations during the conversion.
Note
This feature is only available if your database is Db2 10.1 or higher and if the database is not using the
Db2 pureScale Feature.
Note
This feature is only available if your database is Db2 10.5 FP5SAP2 or higher and if the requirements
regarding component SAP_BASIS are fulfilled (see SAP Note 1513862 ).
● Convert SAP Convergent Invoicing tables from row-organized to column-organized and vice versa.
Note
This feature is only available if your database is Db2 10.5 FP5SAP2 or higher and if you are using SAP
Contract Accounts Receivable and Payable (FI-CA) Release 617 or higher and have applied the
correction instructions according to SAP Note 2207508 ).
The following table provides an overview of the most important changes to this document:
4.7 June 4, 2018 Document update due to new DB6CONV Version 6.40
Section Performing a Database-Level Conversion [page 14] was updated: new op
tion Correct Inline Length of LOB Columns)
Section Available Information on the Conversion Overview Screen [page 39] was
updated: As of Db2 11.1, progress details for copy & replay phase are displayed.
4.6 February 5, 2018 Document update due to new DB6CONV Version 6.39
4.5 March 9, 2017 Section about database-level conversions was updated (new option Drop Hidden
Columns)
4.4 August 4, 2016 Document update due to new DB6CONV Version 6.30
New section in Appendix (overview of SAP BW Tables Considered for BLU Accelera
tion)
4.3 March 1, 2016 Document update due to new DB6CONV Version 6.24
DB6CONV now supports tablespace pools. For more information about tablespace
pools, see also SAP Note 2267446 .
4.2 November 6, 2015 Document update due to new DB6CONV Version 6.20
4.11 February 17, 2015 Document layout changed (new cover page and font style); content unchanged
4.1 December 9, 2014 Document update due to new DB6CONV Version 6.00; BW conversion is now pos
sible
4.0 November 22, 2013 Document update due to new DB6CONV Version 5.30:
3.0 July 20, 2012 Document update due to new DB6CONV Version 5.20
2.1 March 26, 2012 Document update due to new DB6CONV Version 5.15:
2.0 April 15, 2011 Document update due to new DB6CONV Version 5.07:
Usage Requirements
● IBM Db2 for Linux, UNIX, and Windows Version 9.7 or higher
● SAP Basis Release 7.00 and higher
For conversions with Db2 releases lower than Db2 Version 9.7 and SAP systems lower than 7.00, you can only
use report DB6CONV version 5.36 (or higher) as described in SAP Note 2091519 .
Constraints
● Report DB6CONV is not part of the standard SAP delivery. The latest version is available as attachment to
SAP Note 1513862 .
The attached archive DB6CONV_V<version>_<timestamp>.SAR contains the data- and the co-file of the
corresponding DB6CONV transport that you have to import into the SAP system.
The transport consists of the following objects:
○ Reports DB6CONV, DB6CONVP, DB6CONV_BW, and DB6CONV_CI
○ Include DB6CONV_ITC_SUB
○ Tables DB6CONVT, DB6CONVL, DB6CONVQ, DB6CONVE, and DB6CONVO
● Report DB6CONV can only move tables of the AS ABAP. Moving other kinds of tables, for example, of an
SAP Java system, is not supported by DB6CONV.
● Report DB6CONV cannot perform explicit structure changes when moving a table.
Implicit structure changes to tables, such as the conversion of LONG VARCHAR columns to VARCHAR
columns or LONG VARCHAR columns to columns of type LOB, are performed depending on the release
level of the ABAP Dictionary and the kernel database shared library dbdb6slib.
IBM Db2 Version 11.1 for Linux, UNIX, and Windows Db2 11.1
IBM Db2 Version 10.5 for Linux, UNIX, and Windows Db2 10.5
IBM Db2 Version 10.1 for Linux, UNIX, and Windows Db2 10.1
IBM Db2 Version 9.7 for Linux, UNIX, and Windows Db2 V9.7
● Tablespace conversion
● Database-level conversion
● External conversion
● BW conversion
● Conversion of SAP Convergent Invoicing (SAP CI) tables
Procedure
Context
The following procedure applies to the conversion of a single table as well as to tablespace conversion because
most parameters to be specified are identical on the Single Table and the Tablespace tab pages:
Procedure
1. In the field Table to Be Converted or Tablespace to Be Converted, enter a name or choose the table or
tablespace to be converted from the dropdown list. As of DB6CONV V6.24, you also find tablespace pools
in the list of available source tablespaces. You can recognize them by the curly brackets in the name. If
there are warning messages regarding a tablespace pool, the pushbutton Display Messages appears. Make
sure you click it and read the info.
3. If you want to perform a conversion using a target data class, choose the radio button Use Target Data
Class, and in the field Target Data Class, choose the appropriate target data class for the table or
tablespace to be converted from the dropdown list. This dropdown list not only provides all existing data
classes, but also their respective data and index tablespaces.
4. For Db2 10.1 or higher and if the database is not using the DB2 pureScale Feature:
a. In the Insert Time Clustering field, choose the appropriate option for the table or tablespace to be
converted from the dropdown list. The default value is Keep Current Status, that is, the cluster attribute
of the table or of all tables in a tablespace is kept.
b. Optional and only available for single table conversions: If you want to display the ITC characteristics of
the table and check or modify the choice that you made in the previous step a., choose the ITC Status
pushbutton.
5. Only for tablespace conversions:
a. Always create the conversion queue first using the Create Queue pushbutton.
b. Then choose the Display/Edit Queue pushbutton to display and maintain the list of tables that are part
of the tablespace conversion, or choose the Save pushbutton to directly save the conversion. In the
Display/Edit Queue dialog box, you can view and maintain the queue of tables to be converted as
described in Maintaining the Conversion Queue of a Mass Conversion [page 24].
○ Use COPY_WITH_INDEXES
With COPY_WITH_INDEXES, all indexes of the target table are created before the data transfer. By
default, this option is preselected in DB6CONV.
If you deselect this option, all indexes of the target table are created after the data transfer. This may
result in a shorter total time for the data transfer itself, but in a higher temporary space request for the
creation of the indexes. In addition, you might run into LOG_DIR full errors (SQL error SQL0964C). This
is particularly critical in HADR scenarios.
COPY_WITH_INDEXES is an option of the ADMIN_MOVE_TABLE procedure, which is called up by
DB6CONV for the online table move. As opposed to the ADMIN_MOVE_TABLE procedure, DB6CONV uses
COPY_WITH_INDEXES as its default to avoid LOG_DIR full errors.
○ Use LOAD
If you select this option, the data transfer is done using the LOAD command.
○ RECOVERABLE
The data transfer is logged. To use this option, you must specify the path for corresponding log
files for the conversion option COPY YES TO and for the profile LOAD (see Specifying Conversion
Options [page 28]).
○ NON-RECOVERABLE
The data transfer is not logged. A backup after the conversion is required.
Note
The option REDIRECT is not available for BW conversions from row-organized to column-organized
(or vice versa) and for conversions of SAP Convergent Invoicing tables.
Note
The option Stop Before Switch is only available for the conversion of a single table.
Note
Do not select this option if, for example, you want to compare the source table with the target table
after the conversion. In this case, you have to manually drop the source table at a later point in time.
Note
For BW conversions from row-organized to column-organized (and vice versa) and for conversions of
SAP Convergent Invoicing tables, you cannot specify compression options.
By default, compression for tables and indexes is set to the value KEEP, that is, the compression after the
conversion will be the same as before the conversion.
8. Save your entries.
The Conversion Overview screen appears and the new conversion job is displayed with status PLANNED in
the Conversion Jobs area.
9. Optionally, you can check and maintain the conversion parameters and the conversion queue as described
in Maintaining the Conversion Parameters [page 23] and Maintaining the Conversion Queue of a Mass
Conversion [page 24].
10. Start the conversion job (see Starting a Conversion Job [page 29]).
The different types of database-level conversions are described in the following. Depending on your database
version, the available scenarios vary.
To simplify the task of converting a set of tables that even may reside in different tablespaces, you can write
their names to a file and upload the file into DB6CONV.
As of DB2 V9.7, inline LOBs were introduced providing faster access to data than LONG VARCHAR columns.
During a LONGLOB conversion with report DB6CONV, new tables are created with inline LOB columns instead
of LONG VARCHAR columns. During the LONGLOB conversion, all tables with LONG VARCHAR columns are
collected in the conversion queue and converted.
To perform a LONGLOB conversion, go the Database-Level tab page and choose the radio button Replace LONG
VARCHAR Columns by LOB Columns (LONGLOB Conversion). Then proceed as described in step 6 to 13 in
Converting a Table or a Tablespace [page 10].
Note
This option is only available as of DB2 10.1 and if the database is not using the Db2 pureScale Feature.
DB6CONV can create a list of tables that might be candidates for a conversion to ITC tables. For information
about how ITC candidates are chosen, see Insert Time Clustering (ITC) Tables and Candidates [page 37].
To convert ITC candidates to ITC tables, go to the Database-Level tab page and choose the relevant radio
button. Then proceed as described in step 6 to 13 in Converting a Table or a Tablespace [page 10].
Note
This option is only available as of Db2 10.1 and if the database is not using the Db2 pureScale Feature.
To reconvert ITC tables to regular tables, go to the Database-Level tab page and choose the relevant radio
button.
Note
This option is only available as of Db2 10.1.
In some scenarios, it might be necessary to work with hidden columns in a table that need to be dropped later.
This is the case, for example, when an SAP system upgrade with the zero-downtime option (ZDO) in the
software update manager (SUM) is reverted. Instead of directly dropping one or more columns of a table
(which will put the table in a "reorg pending" state), these columns are altered to “implicitly hidden”. This has
the advantage that the columns are invisible and the table is still available. To get rid of the hidden fields
eventually, the table needs to be reorganized. Alternatively, to avoid the downtime for a reorganization, you can
convert the tables using DB6CONV with the conversion type Drop Hidden Columns.
During such a conversion, all tables with hidden columns are collected in the conversion queue and converted,
which implicitly recreates each table without the hidden columns.
To perform this kind of conversion, go to the Database-Level tab page and choose the radio button Drop Hidden
Columns. Then proceed as described in steps 6 to 13 in Converting a Table or a Tablespace [page 10].
For Db2 V9.7 or higher and SAP Kernel 7.00 or higher, BLOB and CLOB columns are created with a default
INLINE LENGTH clause as described in SAP Note 29824 . If the existing inline length of a BLOB or CLOB
column is less than its default value, it may lead to bad performance when this data is accessed.
During a conversion with option Correct Inline Length of LOB Columns, all tables of this kind are collected in the
conversion queue and converted, which implicitly corrects the inline length of their BLOB and CLOB columns to
the default value.
To perform this kind of conversion, go the Database-Level tab page and choose the radio button Correct Inline
Length of LOB Columns. Then proceed as described in step 6 to 13 in Converting a Table or a Tablespace [page
10].
Procedure
1. On the New Conversion screen, go to the Database Level tab page and choose one of the above-mentioned
options.
Note
Selecting a target long tablespace is optional whereas the selection of a target index tablespace is
mandatory unless you have chosen a tablespace pool as target data tablespace. If you have chosen a
tablespace pool as target data tablespace, leave the Target Index Tablespace and Target Long Tablespace
columns empty because they are automatically determined and any manual selections would be
overwritten.
Context
As of DB6CONV Version 6.20 with Db2 10.5 FP5SAP2 and if the SAP Business Warehouse requirements are
fulfilled (see SAP Note 1513862 ), you can perform BW conversions using the Business Warehouse tab page.
You can convert SAP BW object tables from row-organized to column-organized and vice versa, or without
changing the table organization. You can also combine BW conversions with a move of the affected tables to
new tablespaces. If BW object tables are converted from row-organized to column-organized or vice versa, the
additional task of BW Post-Processing is executed per BW object.
● Aggregates for InfoCubes that are being converted from row-organized to column-organized are disabled.
● BW metadata for the converted BW objects is updated.
To check the consistency of BW objects or to repeat the post-processing for BW objects, you can also execute
BW Post-Processing Only without converting the related SAP BW object tables.
Note
When the RSADMIN parameter DB6_ALLOW_CDE is set to NO, the creation of new objects for BLU
Acceleration is blocked in the system. In this case, the DB6CONV report informs you that conversion to BLU
Acceleration is not allowed, and the target BLU Acceleration is grayed out on the Business Warehouse tab
page.
For more information, see the guide SAP Business Warehouse on IBM Db2 10.5 and Higher for Linux, UNIX, and
Windows: Administration Tasks at https://help.sap.com/viewer/db6_bw.
Procedure
1. On the New Conversion screen, choose the Business Warehouse tab page and select/deselect the BW
object types that you want to convert, that is, InfoCubes, DataStore objects (DSO), InfoObjects, or
Persistent Staging Area (PSA) tables. By default, all types are selected.
2. Select the BW objects to be converted by name in one of the following ways:
If you want to include InfoObjects in the conversion queue that are referenced in the InfoCubes and
DataStore Objects to be converted, select the Include Dependent InfoObjects checkbox. By default, this
option is preselected because the reporting performance usually improves if the InfoObject tables that are
referenced in the queries have the same layout as the InfoCube or DSO tables. If you do not want to convert
InfoObjects, deselect this option.
Note that you cannot change the table organization of flat InfoCubes (Flat ICubes) as flat InfoCubes are
always column-organized. However, if you select dependent InfoObjects of a flat InfoCube, these
dependent InfoObjects can be converted (to BLU Acceleration or to row-organization).
○ Depending on how you want to convert your selected BW object tables, choose the Target: BLU
Acceleration radio button or the Target: Row-Organized radio button in the Conversion section.
○ If you want to convert the selected BW object tables without changing the table organization, choose
Keep Table Organization.
If necessary, you can now modify the queue by explicitly deleting one or more BW objects before saving the
conversion.
By double-clicking a BW object, you can display a list of all tables belonging to this object that will be
converted. As of DB6CONV 6.30, this list tells you the current (that is, source) table organization, the
calculated target table organization as well as what kind of conversion will be performed.
Note
Note that the target state BLU Acceleration of a BW object might consist of a mix of row- and column-
organized tables because not all tables of a SAP BW object can be converted to column-organized. To
find out which tables can be column-organized, see SAP BW Tables Considered for BLU Acceleration
[page 44] in the appendix.
Therefore, as of DB6CONV 6.30, if you have tables that are not eligible to BLU Acceleration, but are
nevertheless column-organized (“C “), a BW conversion with Target: BLU Acceleration will convert them
back to row-organized (“R “).
6. To specify a different set of target tablespaces, use the Target Tablespaces button on the New Conversion
screen. By default, the tables of a BW conversion are converted within their own tablespaces. The Choose
Target Tablespaces screen displays all tablespaces that contain tables of the conversion queue in the
Source Data Tablespace column. If you choose a tablespace from a dropdown list in the Target Data
Tablespace, Target Index Tablespace, or Target Long Tablespace columns, all tables currently residing in the
source tablespace will be moved to the newly selected set of target tablespaces during the conversion.
As of DB6CONV 6.30, if you choose new tablespaces, all tables belonging to the BW object are converted,
even if their table organization does not change. If you don’t choose new tablespaces, only those tables are
converted where the table organization changes during the conversion.
Note
Selecting a target long tablespace is optional whereas the selection of a target index tablespace is
mandatory unless you have chosen a tablespace pool as target data tablespace. If you have chosen a
tablespace pool as target data tablespace, leave the Target Index Tablespace and Target Long Tablespace
columns empty because they are automatically determined and any manual selections would be
overwritten.
After saving, all BW conversions are named <BW>_#, where # is a 6-digit ascending number starting with
<BW>_000001 for the first BW conversion.
Prerequisites
You are using SAP Contract Accounts Receivable and Payable (FI-CA) Release 617 or higher running on Db2
10.5 FP5SAP2 or higher.
1. Check and implement the prerequisites for column-based storage (BLU Acceleration) from SAP Note
1819734 .
2. Apply SAP Note 2207508 .
Context
As of version 10.5, IBM Db2 for Linux, UNIX, and Windows supports column-based storage for the analysis of
unbilled items in SAP Convergent Invoicing. If you want to use column-based storage, you need to convert the
relevant database tables from a row-based to a column-based table layout. For more information, see SAP
Note 2136479 .
On the Convergent Invoicing tab page, you can convert SAP Convergent Invoicing tables from row-organized to
colum-organized and vice versa. You can also combine these conversions with the move of the affected tables
to new tablespaces.
After you have converted tables to column-based, new tables that are created for a billing class will
automatically be created in column-based storage.
Procedure
1. On the New Conversion screen, choose the Convergent Invoicing tab page.
Note
Selecting a target long tablespace is optional whereas the selection of a target index tablespace is
mandatory unless you have chosen a tablespace pool as target data tablespace. If you have chosen a
tablespace pool as target data tablespace, leave the Target Index Tablespace and Target Long Tablespace
columns empty because they are automatically determined and any manual selections would be
overwritten.
After saving, all mass conversions of SAP Convergent Invoicing tables are named <CI>_#, where # is a 6-
digit ascending number starting with <CI>_000001 for the first mass conversion.
In DB6CONV, you can also run conversion jobs for SAP BW that repair distribution keys.
These conversion jobs are generated by tools outside the DB6CONV report and are then listed in DB6CONV as
object type External. The object name of the corresponding conversion job starts with <EXT>_.
Conversion Task Name of Conversion Job Tool to Generate the Con Relevant SAP Notes
version Job
For more information, see the database administration guide SAP Business Warehouse on IBM Db2 10.5 and
Higher for Linux, UNIX, and Windows: Administration Tasks at https://help.sap.com/viewer/db6_bw.
For any conversion job, you can display or maintain the conversion parameters as follows:
● For a new conversion, you can specify the parameters for the conversion job on the New Conversion screen
(not applicable for external conversions).
● For conversions with status PLANNED, you can display or maintain the current conversion parameters by
selecting an entry on the Conversion Overview screen and choosing the Details pushbutton.
Exception: For tablespace conversions, you cannot change the option that you have chosen in the Insert
Time Clustering field anymore.
For a mass conversion, you can display or maintain the current queue of tables to be converted as follows:
● For a new conversion, you can display or maintain the initial conversion queue while specifying the
parameters for the conversion job by choosing the Conversion Queue pushbutton on the New Conversion
screen.
● For conversions other than Convergent Invoicing conversions with status PLANNED, you can display or
maintain the current conversion queue by choosing the Details pushbutton on the Conversion Overview
screen and then the Conversion Queue pushbutton on the Details for <object name> screen.
● For Convergent Invoicing conversions or conversions with any other status than PLANNED, you can only
display the conversion queue by choosing the Details pushbutton on the Conversion Overview screen and
then the Conversion Queue pushbutton on the Details for <object name> screen.
For tablespace conversions, all tables that reside in the specified tablespace and that have the table schema of
the AS ABAP system are initially displayed in the conversion queue in alphabetical order. A message window is
displayed if the tablespace contains tables with a table schema other than the table schema of the AS ABAP
system.
For LONGLOB conversions, all tables that have the table schema of the AS ABAP system with at least one
LONG VARCHAR field are initially displayed in the conversion queue in alphabetical order independently of their
tablespace location.
For ITC conversions, all ITC candidate tables are initially displayed in the conversion queue in alphabetical order
independently of their tablespace location.
Similarly, for the reconversion of ITC tables to regular tables, all ITC tables are initially displayed in the
conversion queue in alphabetical order independently of their tablespace location.
For BW conversions, all BW objects of the selected object types of which at least one table is not in the target
table organization are initially displayed in the conversion queue.
The following information is displayed in the Conversion Queue dialog box (not for BW conversions; see the end
of this chapter for information about what is listed in the conversion queue for BW conversions):
Column Description
Note
Without current statistics, the real size of a table can de
viate from the size specified here.
For Db2 10.1 or higher and if the database is not using the Db2 pureScale Feature:
Column Description
Possible values:
YES:
NO:
KEEP:
If you need detailed information about ITC candidates, you can display the following columns by choosing the
ITC Status pushbutton:
Column Description
For more information, see Insert Time Clustering (ITC) Tables and Candidates [page 37].
The following information is displayed in the Conversion Queue dialog window for BW conversions only:
If you click on a BW object in the list, a dialog window opens displaying the following information for the tables
related to the BW object:
Column Description
Note
Without current statistics, the real size of a table can de
viate from the size specified here.
In the Conversion Queue dialog box, you can perform one of the following actions by choosing the appropriate
pushbutton:
You can display and maintain conversion options on the Conversion Overview screen. The conversion options
will apply to all conversions that you create and that use the LOAD command from IBM Db2.
Context
DB6CONV uses the Db2 command LOAD to load data into tables. With the conversion options, you can define
additional options for the LOAD command.
Procedure
COPY YES With the COPY YES option, you define the target location
where data is stored during a recoverable load. The values
that you can enter are:
○ USE TSM
○ TO <device/directory>
○ LOAD <lib-name (including full
path)>
Note
In an HADR scenario, make sure that the standby can
access the files at the location that you have specified.
CPU_PARALLELISM For conversions defined with Use LOAD, you can select
and specify one or more of these options to finetune the
DATA_BUFFER
DB2 LOAD command.
DISK_PARALLELISM
FETCH_PARALLELISM
SORT BUFFER
For more information about the DB2 LOAD command, see the IBM Knowledge Center at https://
www.ibm.com/support/knowledgecenter/en/SSEPGG_11.1.0/com.ibm.db2.luw.admin.cmd.doc/doc/
r0008305.html .
Procedure
1. On the Conversion Overview screen, start the conversion job by selecting the planned job from the list and
choosing the Start pushbutton.
2. In the Start Conversion dialog box, specify the following parameters:
○ The number of parallel jobs used during the conversion (only available for mass conversions)
The value can range from 1 up to the number of batch jobs configured in the system.
○ If you want the conversion job to start immediately, choose the radio button Start Batch Job(s)
Immediately.
○ If you want the conversion job to start at a specified time, choose the radio button Schedule Batch
Job(s).
3. Confirm your entries by choosing the OK pushbutton.
If you specified a certain date and time for the conversion to start, the status of the conversion job changes
from PLANNED to SCHEDULED on the Conversion Overview screen. As soon as the conversion starts, the
status changes from SCHEDULED to RUNNING.
If you specified the conversion to start immediately, the status of the conversion job changes from
PLANNED to RUNNING.
You can monitor the progress of a running conversion using the information displayed in the Comment field on
the Conversion Overview screen.
Note
Since there is no automatic refresh, make sure that you manually refresh the information to always have the
most current information displayed.
The Comment field contains the number of the conversion step that is currently being executed. For the data
transfer step, the number and percentage of currently transferred table rows is also displayed.
where
<x> stands for the number of tables that were already successfully converted
<y> stands for the total number of tables in the conversion queue at the start of the mass conversion
<z> stands for the size of data in KB that was successfully moved
BW Conversion
where
<x1> stands for the number of tables that were already successfully converted
<y1> stands for the total number of tables in the conversion queue at the start of the mass conversion
<z> stands for the size of data in KB that was successfully moved
<x2> stands for the number of BW objects that were already successfully converted, that is, all tables
belonging to this BW object have been successfully converted
<y2> stands for the total number of BW objects in the conversion queue at the start of the mass conversions
The tables that are implicitly converted during a mass conversion are displayed in the Tables from Mass
Conversions area including their status.
For a BW conversion, an additional entry with the name of the BW object and the comment BW Post Processing
is displayed in the Tables from Mass Conversions area after the last table belonging to this BW object was
successfully converted.
If you doubleclick such an entry (or select it and choose the Conversion Log button), the detailed results of the
BW post-processing for the selected BW object is displayed. If the status is ERROR, the BW object is not
consistent. Check the results and contact SAP support, if necessary.
If a subsequent BW post-processing step for the same BW object succeeds, the status of older erroneous
entries changes to ERROR (RESOLVED) and those entries can be deleted from the Tables from Mass
Conversions list.
As of Db2 11.1, you can use the Conversion Overview Screen also to get detailed information about the copy &
replay phase of a conversion.
Related Information
Once you have planned and started a conversion job, you can perform the following actions:
Pausing a Conversion
Note
This pushbutton is only available for mass conversions with status RUNNING.
You can interrupt a mass conversion with status RUNNING by selecting the respective conversion job and
choosing the Pause pushbutton. All parallel jobs of the selected mass conversion stop after each currently
ongoing table conversion, which is part of the mass conversion has finished. You have to manually reset or
continue the paused conversion later on.
Continuing a Conversion
Note
This pushbutton is only available for aborted single table conversions or aborted or paused mass
conversions.
An aborted single table conversion that was started by a batch job of a parent mass conversion is automatically
continued unless you reset or continue the conversion on table level earlier by selecting the appropriate table
from the list in the Tables from Mass Conversion area and choosing the appropriate pushbutton.
When you continue a mass conversion, you can also change the number of parallel jobs that you specified
when starting the conversion.
Replaying a Conversion
Note
This pushbutton is only available for conversions of single tables with status PAUSED.
For conversions of a single table that were defined with the Stop Before Switch option, the conversion is
stopped after the initial data copy, and the conversion status is set to PAUSED. Continuing a paused conversion
will at first apply data modifications of the original table to the target table (replay) and switch the table
afterwards. The more data there is to be replayed, the longer it takes until the table is finally switched.
Therefore, we recommend that you perform regular replay actions using the Replay pushbutton before
continuing a conversion.
Aborting a Conversion
Note
This pushbutton is only available for single table conversions with status RUNNING.
You can abort the conversion of a single table by either selecting an entry in the Conversion Jobs area or in the
Tables from Mass Conversion area and choosing the Abort pushbutton.
We recommend that you only use this option in emergency cases, for example, if a conversion is running too
long.
Resetting a Conversion
Note
This pushbutton is only available for aborted single table conversions or aborted or paused mass
conversions.
For a single table conversion, resetting a table restores the state of the table from before the conversion if the
table has not yet been renamed.
Resetting a mass conversion simply empties the queue of the marked mass conversion and resets table
conversions with status ABORTED that belong to this mass conversion. Then, all tables that have already been
successfully converted reside in the target tablespace. All tables that have not yet been converted remain in the
source tablespace.
Note
Be aware that a conversion reset is always executed immediately and with only one parallel process. This
means that you cannot schedule a reset job to start at a specific time.
Note
This pushbutton is only available for currently running mass conversions.
You can change the number of parallel jobs of a running mass conversion dynamically without stopping it by
selecting the appropriate conversion job from the list and choosing the +/- Jobs pushbutton.
You might want to change the number of parallel jobs, for example, in the following situations:
You can display log information about any conversion job independently of whether it is running or whether it
has finished. To do so, select the appropriate entry on theConversion Overview screen and choose the
Conversion Log pushbutton. Alternatively, you can double-click the appropriate entry:
● If you doubleclick an entry in the Tables from Mass Conversion area or an entry with object type Table in the
Conversion Jobs area, the conversion log for the conversion of this particular table is displayed.
● If you doubleclick an entry of object type Tablespace or LONGLOB in the Conversion Jobs area, a summary
log is displayed for this mass conversion containing information about the conversion queue and the batch
jobs.
● If you doubleclick a line containing Conversion Job <batch job name> in the summary log for this mass
conversion, you switch to the summary log of that batch job that contains information about the single
Note
This pushbutton is only available for conversions with status FINISHED, RESET, or PLANNED.
By choosing the Delete pushbutton, you can delete the detailed conversion logs of a single or multiple selected
conversions.
If a single table conversion (either listed in the Conversion Jobs section or in the Tables from Mass Conversions
section) aborts and you want to reproduce the error using the trace option of the ADMIN_MOVE_TABLE
procedure, you can do so by selecting the single table conversion and entering AMT1 into the transaction field
of the SAP GUI. A message window appears to confirm that the trace has been switched on. If you continue
with the selected table conversion, trace files are now written to the Db2 dump directory.
In the following, the steps and relevant actions of a table conversion are described.
Starting with DB6CONV V6.20, all conversions are performed using the SYSTOOLS.ADMIN_MOVE_TABLE
stored procedure (in the following referred to as AMT). For more information, see SAP Note 1543745 .
Step 0
Step 1
● Calls the AMT stored procedure for the table to be converted and the target tablespaces with operation
parameter COPY.
● Copies all data from the source table to the target table.
You cannot monitor the progress of the data transfer in the detailed conversion log but in the Comment
field on the Conversion Overview screen.
Step 3
● Calls the AMT stored procedure for the table to be converted and the target tablespaces with operation
parameter REPLAY.
Note
If the conversion has been defined with the Stop Before Switch option, step 4 is subsequently not
executed, and the status of the conversion job changes from RUNNING to STOPPED.
● Updates the statistics for the table to be converted if necessary, that is, if SYSCAT.TABLES-CARD is less or
equal zero.
Step 4
● Calls the AMT stored procedure for the table to be converted and the target tablespaces using operation
parameter SWAP.
As a result, the source table is switched to the target table and the temporary objects that were created
during step 1 are deleted.
● Changes the data class for the target table if you have specified a target data class. For virtual tables, this is
the only action that is executed, the rest of step 4 is skipped.
● Deletes the original table if you selected Delete Source Table After Conversion when defining the conversion
job.
● Displays again detailed information about the table size and compression.
● Displays again the structure of the target table.
As of Db2 10.1, you can create insert time clustering (ITC) tables. ITC tables are based on the infrastructure for
MDC tables and allow you to reclaim free space without reorganization using the REORG command with the
RECLAIM EXTENTS ONLY clause. In an ITC table, records are clustered based on their insert time.
For an effective reclamation of free space for ITC tables, complete extents of the table must be empty.
Therefore, a regular table should only be considered an ITC candidate if records that are inserted together are
also deleted together. In an SAP system, this access pattern is typical for tables from which data is archived.
For example, a business object contract has an associated archiving object. During archiving, usually all
contracts up to a specific creation date are written to the archive and afterwards deleted from the
corresponding tables.
Note
The fact that data is archived from a table does not necessarily mean that there are empty extents after
archiving. Often, several criteria can be used to decide which records should be archived. These criteria
might not be based on the creation time.
Based on the actual archiving session (see SAP transaction SARA Statistics ) in the system, DB6CONV
can create a list of tables that might be candidates for a conversion to ITC tables. DB6CONV suggests tables
that fulfill all of the following conditions:
● The tables are part of an archiving object for which at least one completed archiving session exists. A
complete archiving session exists if data was archived and afterwards deleted from several or all tables
that belong to the archiving object.
● The tables are not already ITC or MDC tables.
● The tables are at least 1 GB in size.
In the Conversion Queue dialog box, the following information is displayed for the ITC candidates:
Note
It is not guaranteed that deletions take place from all tables of an archiving object. Therefore, the list of ITC
table candidates can contain tables where actually no deletions took place. These tables, however, are
typically less than 1 GB in size.
Multidimensional and insert time clustering extent management in the IBM Knowledge Center at https://
www.ibm.com/support/knowledgecenter/SSEPGG_10.1.0/com.ibm.db2.luw.admin.partition.doc/doc/
c0054692.html
● Conversion jobs
● Tables from mass conversions
● Progress details for copy & replay (as of Db2 11.1 and higher)
Here you can see all the conversion jobs that are known to DB6CONV. A conversion job is a conversion that you
have explicitly defined using DB6CONV or that has been defined outside DB6CONV using a BW function
module.
Column Description
Possible values:
● PLANNED
● SCHEDULED
● PAUSED
● RUNNING
● ABORTED
● FINISHED
● RESET
● NEW
● PRELIMINARY
The conversion job statuses NEW and PRELIMINARY reflect the job status of the correspond
ing batch job as shown in SM37. They may occur if a batch job was planned, but could not be
started. Conversions with this status cannot directly be continued or reset. In these cases,
you have to delete the corresponding batch job in SM37. After refreshing the conversion over
view, the status of the conversion job should have switched to ABORTED.
Possible values:
● Table
● Tablespace
● DLCONV
● External
● BW
● CI
End Date Date and time when the status of a running conversion job changed to one of the following
statuses:
End Time
● PAUSED
● ABORTED
● RESET
● FINISHED
Duration Total runtime of the conversion job since its start (does not include wait times)
This does not include LOB data that are not inlined.
A value of 1- indicates that no statistics exist for the corresponding table. Therefore, its size is
unknown.
● For status PLANNED: Total size of all tables of the current conversion queue
● For statuses other than PLANNED: Total size of all tables of the conversion queue at the
time the mass conversion was started.
Jobs Number of currently defined parallel processes for the conversion job
A value of 1 is displayed for single table conversions and for mass conversions with status
PLANNED.
For mass conversions with status PLANNED, you can specify the number of parallel jobs in
the Start Conversion dialog box.
For mass conversions with a status other than PLANNED, the number of parallel jobs can be
changed as follows:
● Manually when you continue an aborted or paused mass conversion (in the Continue
Conversion dialog box)
● Dynamically while the mass conversion is running using the +/- Jobs pushbutton on the
Conversion Overview screen
For more information, see Available Actions on the Conversion Overview Screen [page 31].
Comment Information that is related to the conversion as such or to a specific conversion step
For single table conversions with status RUNNING, the current conversion step is displayed.
While data is being transferred, the progress of the data transfer is also displayed.
For tablespace and database-level conversions, the progress of the conversion is displayed
as follows:
where
<x> stands for the number of tables that have already been successfully converted
<y> stands for the total number of tables in the conversion queue at the start of the mass
conversion
<z> stands for the size of data in KB that was successfully moved
For BW conversions, the Comment field displays information based on the following formula:
where
<x1> stands for the number of tables that were already successfully converted
<y1> stands for the total number of tables in the conversion queue at the start of the mass
conversion
<z> stands for the size of data in KB that was successfully moved
<x2> stands for the number of BW objects that were already successfully converted, that is,
all tables belonging to this BW object have been successfully converted
<y2> stands for the total number of BW objects in the conversion queue at the start of the
mass conversions
All table conversions are listed that have implicitly been started by the batch process that you defined for the
parent mass conversion job. The mass conversion job itself is displayed in the Conversion Jobs area.
The following information is displayed in the Tables from Mass Conversions area:
Column Description
Possible values:
● PLANNED
● SCHEDULED
● PAUSED
● RUNNING
● ABORTED
● FINISHED
● RESET
● ERROR (only for BW post processing)
● ERROR (RESOLVED) (only for BW post processing)
● NEW
● PRELIMINARY
The conversion job statuses NEW and PRELIMINARY reflect the job status of the corre
sponding batch job as shown in SM37. They may occur if a batch job was planned, but could
not be started. Conversions with this status cannot directly be continued or reset. In these
cases, you have to delete the corresponding batch job in SM37. After refreshing the conver
sion overview, the status of the conversion job should have switched to ABORTED.
End Date
End Time
Duration Total runtime required for a conversion job with status RUNNING (does not include wait
times)
This does not include LOB data that are not inlined.
-1 indicates that no statistics exist for the corresponding table. Therefore, its size is un
known.
Parent Name of the batch process of the mass conversion job that implicitly performed the conver
sion of a single table
Comment Information that is related to the conversion as such or to a specific conversion step
For single table conversions with status RUNNING, the current conversion step is displayed
as well as the progress of the data transfer.
Information Displayed in the Progress Details for Copy & Replay Area (Db2
11.1 or higher)
All table conversions are listed with AMT Status Copy or Replay. Especially for large tables where the related
Conversion Steps [page 35] no. 2 and 3 may take a long time, detailed information about the copy or replay
performance can be helpful. For more information about the displayed data see ADMIN_MOVE_TABLE
procedure - Move tables online in the IBM Knowledge Center at https://www.ibm.com/support/
knowledgecenter/en/SSEPGG_11.1.0/com.ibm.db2.luw.sql.rtn.doc/doc/r0055069.html .
The following information is displayed in the Progress Details for Copy & Replay area:
Column Description
Throughput [rows/s] Indicates the copy performance based on the following formula:
COPY_TOTAL_ROWS/(MIN(CURRENT_TIMESTAMP, COPY_END)-COPY_START)
Inflow Staging [rows/s] Number of rows per second that are inserted into the staging table
Outflow Staging [rows/s] Number of rows per second that are removed from the staging table
If the conversion has the status PAUSED or ABORTED, n.a. (not available) is returned.
The table provides you with an overview which tables of SAP BW objects can be converted to column-organized
table layout or not.
The table lists the most relevant tables of each SAP BW object and whether they can be column-organized. For
some SAP BW objects such as InfoObjects, only some of their tables can be converted to column-organized
while the rest of the tables remain row-organized. For other SAP BW objects, you can control with RSADMIN
parameters whether tables are converted to row-organized or to column-organized. Note that this is intended,
as not every table conversion to column-organized really results in a performance benefit. The defaults and
possible settings listed here are considered by report DB6CONV as of version 6.30 when you convert tables
from row-organized to column-organized and vice versa.
/.../A...00 Yes
/.../B... No
/.../A...40 No
/.../9D... No
/.../S... Yes
/.../X... Yes
/.../Y... Yes
Other No
/BIC/B... Yes
Other No
Are you looking for more documentation? Here's an overview of information sources that are available for SAP
systems on IBM Db2 for Linux, UNIX, and Windows.
Documentation by SAP
The following documentation is available on SAP Help Portal and SAP Support Portal :
Recommendation
For central access to all our documentation, you can also use our SAP on IBM Db2 overview page on SAP
Help Portal at https://help.sap.com/viewer/p/DB6.
Installation guides for SAP Application Server systems based on SAP NetWeaver:
https://help.sap.com/viewer/30839dda13b2485889466316ce5b39e9/
CURRENT_VERSION/en-US/c8ed609927fa4e45988200b153ac63d1.html
https://support.sap.com/en/tools/software-logistics-tools.html
https://help.sap.com/viewer/30839dda13b2485889466316ce5b39e9/
CURRENT_VERSION/en-US/14678bf6e9a04f4f9dc3a1620e96c0b7.html
Check out and participate in our SAP community for IBM Db2 at https://www.sap.com/community/topic/db2-
for-linux-unix-and-windows.html . Here you'll find blogs, Q&As, whitepapers, videos, and guides. You can also
post blogs and questions and search the community archive for more.
You can find product documentation on your Db2 version in the IBM Knowledge Center at http://
www-01.ibm.com/support/knowledgecenter/SSEPGG/welcome . For Db2 11.1, IBM also offers PDF manuals
at http://www-01.ibm.com/support/docview.wss?uid=swg27050624 .
The IBM Knowledge Center also provides links to IBM Redbooks, developerWorks articles, IBM white papers,
and so on.
By following links to IBM Documentation you are leaving the SAP product documentation and entering a site
that is not hosted by SAP. By using the link, YOU AGREE that unless expressly stated otherwise in your
agreements with SAP you are about to access an external webpage which is not part of SAP’s offering:
(i) the content of the linked-to site and any further external site is not product documentation and you may not
infer any product documentation claims against SAP based on this information;
(ii) the fact that SAP provides links to external sites does not imply that SAP agrees or disagrees with the
contents and information provided on such sites. SAP does not guarantee the correctness of the information
provided.
(III) SAP DOES NOT GIVE ANY REPRESENTATION REGARDING THE QUALITY, SAFETY, SUITABILITY,
ACCURACY OR RELIABILITY OF ANY EXTERNAL WEBPAGE OR ANY OF INFORMATION, CONTENT AND
MATERIALS PROVIDED THEREON;
(IV) YOU VISIT THOSE EXTERNAL WEBPAGES ENTIRELY AT YOUR OWN RISK. SAP SHALL NOT BE DIRECTLY
OR INDIRECTLY RESPONSIBLE OR LIABLE FOR ANY DAMAGE OR LOSS CAUSED OR ALLEGED TO BE
CAUSED BY OR IN CONNECTION WITH YOUR USE OF OR RELIANCE ON ANY CONTENT, GOODS OR
SERVICES AVAILABLE ON OR THROUGH ANY SUCH LINKED WEBPAGE.
Hyperlinks
Some links are classified by an icon and/or a mouseover text. These links provide additional information.
About the icons:
● Links with the icon : You are entering a Web site that is not hosted by SAP. By using such links, you agree (unless expressly stated otherwise in your
agreements with SAP) to this:
● The content of the linked-to site is not SAP documentation. You may not infer any product claims against SAP based on this information.
● SAP does not agree or disagree with the content on the linked-to site, nor does SAP warrant the availability and correctness. SAP shall not be liable for any
damages caused by the use of such content unless damages have been caused by SAP's gross negligence or willful misconduct.
● Links with the icon : You are leaving the documentation for that particular SAP product or service and are entering a SAP-hosted Web site. By using such
links, you agree that (unless expressly stated otherwise in your agreements with SAP) you may not infer any product claims against SAP based on this
information.
Example Code
Any software coding and/or code snippets are examples. They are not for productive use. The example code is only intended to better explain and visualize the syntax
and phrasing rules. SAP does not warrant the correctness and completeness of the example code. SAP shall not be liable for errors or damages caused by the use of
example code unless damages have been caused by SAP's gross negligence or willful misconduct.
Gender-Related Language
We try not to use gender-specific word forms and formulations. As appropriate for context and readability, SAP may use masculine word forms to refer to all genders.