Space could be allocated based on the number of records in the data set rather than on the range of keys. Allocating space based on the number of records requires the use of a cross-reference table. When a record is written in the data set, you must note the physical location as a relative block number, an actual address, or as a relative track and record number. The addresses must then be stored in a table that is searched when a record is to be retrieved. Disadvantages are that cross-referencing can be used efficiently only with a small data set; storage is required for the table, and processing time is required for searching and updating the table.