Show TOC

Selection of MDC DimensionsLocate this document in the navigation structure

Use

When selecting MDC dimensions, 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 characteristics determines the cardinality. Therefore, select a dimension with just one or a 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.

    Recommendation

    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 queries (assign the lowest sequence number to the InfoObject that occurs most often in queries).

    • Sort the dimensions according to selectivity (assign the lowest sequence number to the dimension with the most different data records).

      Caution

      Note: At least one block is created for each value combination in the MDC dimension. This memory area is reserved independently 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 memory. This means that the fact tables use an unnecessarily large amount of memory. Performance of table queries also deteriorates, as many pages with not much information must be read.

Example

The size of a block depends on the PAGESIZE and the EXTENTSIZE of the tablespace. The standard PAGESIZE of the fact-table tablespace with the assigned data class DFACT is 16K. Up to Release SAP BW 3.5, the default EXTENTSIZE value was 16. As of Release SAP NetWeaver 7.0, the new default EXTENTSIZE value is 2.

With an EXTENTSIZE of 2 and a PAGESIZE of 16K, the memory area is calculated as 2 x 16K = 32K, which is reserved for each block.

The width of a data record depends on the number of dimensions and the number of key figures in the InfoCube. A dimension key field uses 4 bytes and a decimal key figure uses 9 bytes.

If, for example an InfoCube has 3 standard dimensions, 7 customer dimensions and 30 decimal key figures, a data record needs 10 x 4 bytes + 30 x 9 bytes = 310 bytes. In a 32K block, 32768 bytes/310 bytes could write 105 data records.

If the time characteristic calendar month (0CALMONTH) and a customer dimension are selected as the MDC dimension for this InfoCube, at least 100 data records should exist for each InfoPackage, for each calendar month and for each dimension key of the customer dimension. This allows optimal use of the memory space in the F fact table. In the E fact table, this is valid for each calendar month and each dimension key of the customer dimension.