Browse Prior Art Database

Dataset Organization Which Minimizes I/O Operations

IP.com Disclosure Number: IPCOM000066582D
Original Publication Date: 1979-Mar-01
Included in the Prior Art Database: 2005-Feb-20

Publishing Venue

IBM

Related People

Authors:
Metcalfe, J Temple, NJ [+details]

Abstract

When an editor is being used for a source program, a map or any other dataset, a large amount of dataset I/O usually occurs. This is because the editor normally copies the original dataset to a temporary one (or into storage) at the start of the editing and copies it back for all SAVE requests. This is normally needed because the editor may not update the original dataset as it processes input because the user may decide to discard his changes. This means that all records, whether changed or not, take part in I/O more than once. This problem is aggravated because the user may frequently SAVE the changes made because of the danger of a system failure during the editing, causing all the changes to be lost. The amount of I/O can be significantly reduced if the source dataset is organized as described below.