When a data set is first loaded, the key sequence of data records and their physical order are the same. However, when data records are inserted, control interval splits can occur, causing the data control intervals to have a physical order that differs from the key sequence.

Figure 1 shows how the control interval is split and the index is updated when a record with a key of 12 is inserted in the control area shown in Figure 2.

Figure 1. Control Interval Split and Index Update

1. A control interval split occurs in data control interval 1, where a record with the key of 12 must be inserted.
2. Half the records in data control interval 1 are moved by VSAM to the free space control interval (data control interval 3).
3. An index entry is inserted in key sequence to point to data control interval 3, that now contains data records moved from data control interval 1.
4. A new index entry is created for data control interval 1, because after the control interval split, the highest possible key is 14. Because data control interval 3 now contains data, the pointer to this control interval is removed from the free list and associated with the new key entry in the index. Note that key values in the index are in proper ascending sequence, but the data control intervals are no longer in physical sequence.