Martin A. Goetz

An approach to the organization and structure of data on Bryant Disc File Memory Systems for sorting and performing other data processing functions is presented. The following areas are covered: characteristics of Bryant Disc File Systems on the Bendix G-20 and RCA 301; two proposed “chaining” structures for data; and functions of a Disk File Executive Routine. The concepts for sorting and performing file maintenance processing using the proposed structure and executive routine are discussed. Additionally, it is shown that sorting can be accomplished without the use of disk storage work areas.

Read-backward Polyphase sorting provides more efficient use of the tapes available to a sort than most other sorting techniques. Backward Polyphase produces a continuous merging process from n - 1 tapes where n is the total number of tapes being used in the sorting process. Any of the available presorting techniques may be used in conjunction with the Polyphase merge sort provided that the presort has the capability of producing both ascending and descending strings and distributing the strings on the various tapes as required by the Polyphase Merge.

This paper describes the application of several new techniques for sorting fixed-length records to the problem of variable-length record sorting. The techniques have been implemented on a Sylvania 9400 computer system with 32,000 fixed-length words of memory. Specifically, the techniques sequence variable-length records of unrestricted size, produce long initial strings of data, merge strings of data at the power of T - 1, where T is the number of work tapes in a system, and do not restrict the volume of input data.

A general technique for sequencing unsorted records is presented. The technique is shown to be applicable for the first stage of a generalized sort program (the formation of initial strings) as well as for sorting records within a memory storage (an internal sort). It is shown that given N records in memory storage, records are sequenced using 1+log2 N tests per record, that initial string lengths will average 2N for random input records, and that reading, writing and processing can be accomplished simultaneously if the computer permits such overlap.