R7I332 - Mehrdimensionales Clustering für InfoCubes

SAP Error Message - Details

• Message type: E = Error

• Message class: R7I - InfoProvider

• Message number: 332

• Message text: Mehrdimensionales Clustering für InfoCubes

Was verursacht dieses Problem?

You want to select InfoCube dimensions for multidimensional clustering.

Systemantwort

Multidimensional clustering (MDC) organizes a table physically into
blocks. Each block only contains data records with the same values in
the MDC dimensions. The size of the block is fixed and corresponds to
the EXTENTSIZE of the table space. This has at least two pages. A page
size of 16K necessitates a memory requirement of at least 32K for each
block.
Multidimensional clustering can improve the performance of database
requests. If requests are restricted to MDC dimensions, only those
blocks in the table are read that contain the corresponding dimension
values. When data is accessed, block indexes are used that are much
smaller than the usual row indexes, which means you can search through
them more quickly.
The performance of data manipulations (INSERT, UPDATE, DELETE) is also
improved through the physical organization of a MDC table and minimal
effort required for maintaining the smaller block indexes.
The default secondary indexes for the selected InfoCube dimensions are
no longer created in the ABAP Dictionary. Instead, system-internal block
indexes are generated for these dimensions.
The clustering is applied to all the aggregates of the InfoCube. If an
aggregate does not contain a MDC dimension of the InfoCube, or if all
the InfoObjects of a MDC dimension are created as line item dimensions
in the aggregate, the aggregates are clustered using the remaining MDC
dimensions. Index clustering is used for the aggregate if the aggregate
does not contain any MDC dimensions of the InfoCube, or if it only
contains MDC dimensions.

Wie behebe ich diesen Fehler?

Choose a time characteristic for the time dimension or the column with
the dimension key (DIMID) for the time dimension. Only 0CALMONTH or
0FISCPER can be used as time dimensions. The time characteristic must
exist in the InfoCube.
The time dimension always has the sequence number 1.
Select additional dimensions for multidimensional clustering by
assigning sequential numbers for the dimensions.
When selecting, proceed as follows:
Select dimensions for which you often use restrictions in queries.
Select dimensions with a low cardinality.
The MDC dimension is created in the column with the dimension keys
(DIMID). The number of different combinations in the dimension ch
aracteristics determines the cardinality. Therefore, select a dimension
with either one, or few characteristics and with only a few different
characteristic values.
Line item dimensions are not usually suitable since they normally have a
characteristic with a high cardinality.
If you specifically want to create an MDC dimension for a characteristic
with a low cardinality, you can define this characteristic as a line
item dimension in the InfoCube. This is a deviation from the general
rule that line item dimensions contain characteristics with a very high
cardinality. However, this has the advantage for multidimensional
clustering that the fact table contains the SID values of the
characteristic, in place of the dimension keys, and the database query
can be restricted to these SID values.
You cannot select more than three dimensions, including the time
dimension.
Assign sequence numbers using the following criteria:
Sort the dimensions according to how often they occur in BW queries
(assign the lowest sequence number to the InfoObject that occurs most
often in BW queries).
Sort the dimensions according to selectivity (assign the lowest sequence
number to the dimension with the highest selectivity).
Note: At least one block with a memory requirement of at least 32K is
created for each value combination in the MDC dimension. This memory
area is reserved irrespective of the number of data records that have
the same value combination in the MDC dimension. If there is not a
sufficient number of data records with the same value combinations to
completely fill a block, the free memory remains unused. This is so that
data records with a different value combination in the MDC dimension
cannot be written to the block.
If for each combination that exists in the InfoCube, only a few data
records exist in the selected MDC dimension, most blocks have unused
free memory. This means that the fact tables use an unnecessarily large
amount of memory space. The performance of database requests also
deteriorates since many pages have to be read that do not contain very
much information.

Vorgehensweise für Systemadministratoren

Das System gibt eine Fehlermeldung aus und erlaubt Ihnen nicht, mit dieser Transaktion fortzufahren, bis der Fehler behoben ist.

Fehlermeldungsextrakt aus SAP-System. Copyright SAP SE