Calculating the number of blocks required for the alias index

Use the formula that is shown in Figure 1 to calculate the number of index blocks required. Use the result as the value of B_A in Formula for the RACF database size.

Figure 1. The number of blocks that are required for the alias index

B_A = the number of blocks required for the alias index
    (zero if running application identity mapping in stage=0)

B_A ~ the number of blocks required for alias index sequence set, which is a good
     approximation because the space requirement for higher-level index entries
     is a small fraction of this number

      D_A
B_A ~  -----
      E_A

where:
        D_A = Q + S + U + BB + GG

                                                  4096 x 0.5
        E_A =  -------------------------------------------------------------
                 18 + average user or group name length + 3 + average alias name length

        18 = the length of the alias index entry overhead

         3 = the length of the alias index entry key prefix

Note: The calculation for E_A makes the following assumptions:

The alias index blocks are half full (0.5). If you can extend an existing RACF® data set with the IRRUT400 utility, you can:
- Use the IRRUT200 utility program to determine how full the alias index blocks are on the existing data set
- Replace the 0.5 value with a value that you determine
In this case, you can also consider the compressed name length when you specify the average alias name length.
Each alias entry maps to a single user or group profile. While it is possible for multiple users to share a single UID, or for multiple groups to share a single GID, the implementation is not suggested.
The average alias name length is the average length of your SNAME, UNAME, GID, UID, and IPLOOK values.