This note deals with questions and answers about the MaxDB/liveCache system copy.
1. What is a homogeneous system copy?
For a system copy, the following must apply:
a) The target system is a host that has the same behavior concerning byte swapping as the source host.
b) The version of the target database is the same as or higher than the version of the source database. This requirement applies as of MaxDB Version 7.5.
2. What are big endian and little endian? Which endian do I use?
Big Endian:
the most significant byte is stored in memory at the lowest address, and the least significant byte at the highest address. (The big end comes first.) As an analogy, we say "twenty-four" in English; the more significant number, twenty, comes first.
This is also called most significant byte (MSB) ordering.
Little Endian:
the least significant byte of the number is stored in memory at the lowest address, and the most significant byte at the highest address. (The little end comes first.) As an analogy, we say "fourteen" in English; the less significant number, four, comes first.
This is also called least significant byte (LSB) ordering.
For example, 0xAABBCCDD would be stored as follows:
+--------------------------------------+
| Address | Big-Endian | Little-Endian |
|--------------------------------------|
| 3 | DD | AA |
| 2 | CC | BB |
| 1 | BB | CC |
| 0 | AA | DD |
+--------------------------------------+
When converting to Unicode, the export code page must correspond to
the endianness of the target system. More exactly speaking, the export code page should correspond to the endianess of the machine where the R3load import processes will run.
4103 Little Endian Alpha, Intel X86 (and clones),
X86_64, Itanium (Windows+Linux),
Solaris_X86_64
4102 Big Endian IBM 390, AS/400, PowerPC (AIX),
Linux on zSeries (S/390), Linux on Power,
Solaris_SPARC, HP PA-RISC, Itanium (HP-UX)
To determine the processor architecture on Unix systems the command uname gives information. Please match this with one of the endianness classes above. Note that:
some Operating Systems run on different architectures.
the uname options giving platform information differ between Unices.
different platforms may employ the same processor architecture with different endianness; Itanium is an example, cf the list above.
Here are some examples:
uname uname option output
AIX -p powerpc
HP-UX -m 9000/... , i.e. PA-RISC
HP-UX -m ia64
Linux -m i686
Linux -p x86_64
Linux -p ia64
Linux -p ppc64
Linux -p s390x
OS/390
OSF/1 -p alpha
SunOS -p sparc
SunOS -p i86pc
3. Where do I find information about the homogeneous system copy of a MaxDB database?
You can use backup and recovery to perform a homogeneous system copy of a MaxDB database instance.
To generate a database copy, import the backup of a source system into the target system.
To import a backup from another system, a 'normal' recovery is not sufficient. An 'Instance install' or an 'Recovery with Initialization' must be carried out.
4. What is a heterogeneous system copy?
For a heterogeneous system copy, the following must apply:
a) The processors in the two systems (target system and source system) do NOT use the same byte sorting (little endian -> big endian; big endian -> little endian).
b) On the source host, a different database system is used than on the target host (for example, Oracle -> MaxDB).
c) A heterogeneous liveCache system copy may also occur if the conditions mentioned under a) and b) are not met.
5. Which tool is used to perform a heterogeneous system copy?
In general, you can use the tool R3load for the heterogeneous system copy of an OLTP or BW database.
Exceptions are content server instances and liveCache instances.
Since the content server is not included in the SAP Data Dictionary, you cannot use R3load to perform the system copy. You can use the MaxDB tool loadercli for this system copy.
For a heterogeneous system copy of a liveCache instance, you cannot copy the data structures of the LCA objects. The liveCache data is stored in the APO system or the SCM system, and the liveCache is reinstalled on the target system. Then the data is reloaded from the APO system or SCM system to the liveCache.
6. Where do I find information about the liveCache system copy?
If you want to perform a system copy for an SAP liveCache instance, see Notes 457425 and 632357.
Note 886103 describes the system copy of an SAP SCM system as of Version SCM 4.0.
Note 210564 describes the system copy of an SAP APO system.
7. What do I have to consider after a homogeneous liveCache system copy?
The connection data to the liveCache is stored in the database table DBCON. Since this table is also copied, you must ensure that the connection data is adjusted accordingly for the target system using transaction LC10.
After the Homogenous liveCache copy using backup/restore you need to run the following checks before to release the system to the users::
1) Adapt liveCache table /sapapo/ordmap, if you did the complete
APO system copy in Sync.
2)Check/rename the SQL user in the target liveCache.
8. How do I perform a heterogeneous system copy of a content server?
You use the MaxDB tool loadercli to perform the heterogeneous system copy of a content server.
9. What do I have to consider for a heterogeneous system copy using R3load?
a) Parameter settings in the target system with MaxDB:
- Size of the data cache:
Ensure that a sufficient size is configured for the data cache. The general rule for the productive SAP operation is that the configured data cache size amounts to 1% of the total size of the occupied data area. For the period in which the data is loaded, the cache size must be increased in accordance with the memory that is available on the host.
- Parameter MAXCPU:
During loading in MaxDB, you must set the parameter MAXCPU to the number of processors in the database server.
b) Data export
For the export, the data must be unloaded in accordance with the key sequence.
c) Export/import:
With split tables, use the parallel export. We recommend that the import is performed sequentially if the records were previously unloaded in the key sequence.
d) Index creation
Note that the large indexes must not be loaded in parallel. This ensures that several server tasks can be used for the index creation.
e) Precompiler runtime version
If you use the following precompiler runtime versions:
7.4.03 build 44 or higher AND below 7.4.03 build 49
or
7.5.00 build 39 or higher AND below 7.5.00 build 44
refer to Note 1043735.
10. What do I have to consider for the heterogeneous system copy of a BW system?
To load the data without impairing the system performance, you must sort the data in accordance with the key sequence when unloading using R3load.
Also read the following notes:
954268 Optimization of export: Unsorted unloading
777024 BW3.0 and BW3.1 system copy (supplementary note)
771209 NetWeaver 04: System copy (supplementary note)
888210 NetWeaver 7.0/7.10: System copy (supplementary note)
11. Where can I find more information about a system copy?
For more information about the system copy, see the SAP Developer Network how-to at the following link:
https://wiki.sdn.sap.com/wiki/x/LB4B