Page 120 - Accounting Information Systems
P. 120

CHAPT E R 2        Introduction to Transaction Processing  91


                         FI G U R E
                           2-43     TYPES OF POINTERS

                           Physical Address Pointer

                                       CYL  121
                             Record 1  Surface  05                            Next Record
                                       Record 750
                                                Physical Address

                           Relative Address Pointer

                                          135                                  First record
                                                                                  •
                                                                                  •
                                                                                  •
                                                        Conversion                                      Sequential
                                                        Routine                Record 135
                                                                                                             File
                                     Pointer to 135th                             •
                                     Record in the File                           •

                                                                               Last Record

                          Logical Key Pointer

                                        Key             Hashing              Record
                                        9631            Algorithm            9631




                           A relative address pointer contains the relative position of a record in the file. For example, the
                       pointer could specify the 135th record in the file. This must be further manipulated to convert it to the
                       actual physical address. The conversion software calculates this by using the physical address of the begin-
                       ning of the file, the length of each record in the file, and the relative address of the record being sought.
                           A logical key pointer contains the primary key of the related record. This key value is then con-
                       verted into the record’s physical address by a hashing algorithm.

                       Batch Processing Using Sequential Files

                       The most basic computer-processing configuration is batch mode using sequential file structures. Figure
                       2-44 illustrates this method.
                           Each program in a batch system is called a run. In this example, there is an edit run, an AR file
                       update run, an inventory file update run, and two intermediate sort runs. The entire file or batch of records
                       is processed through each run before it moves to the next run. When the last run finishes processing the
                       batch, the session terminates.
                           A prominent feature of this system is the use of sequential files, which are simple to create and
                       maintain. Although sequential files are still used by organizations for backup purposes, their presence in
                       data processing is declining. This file structure is effective for managing large files, such as those used by
                       federal and state agencies that have a high activity ratio. The activity ratio of a file is defined as the per-
                       centage of records on the file that are processed each time the file is accessed. For example, a federal pay-
                       roll file has an activity ratio of 1:1. Each time the payroll file is accessed (payday), all the records on it
                       are processed because everyone gets a paycheck.
   115   116   117   118   119   120   121   122   123   124   125