Browse Prior Art Database

Data Space Manager

IP.com Disclosure Number: IPCOM000045517D
Original Publication Date: 1983-Apr-01
Included in the Prior Art Database: 2005-Feb-07
Document File: 4 page(s) / 19K

Publishing Venue

IBM

Related People

Dey, RW: AUTHOR [+4]

Abstract

This article describes a data space manager (DSM), which is a facility for controlling the operation of a computing system to define, delete, and extend data sets, thus permitting a user to concentrate on definitions of tables (constructs pertinent to the end user's needs) without being distracted by concerns about the characteristics of the direct access storage device (DASD) space in which the tables will reside CREATE and DROP TABLE commands are enhanced to automatically issue CREATE or DROP TABLESPACE commands which in turn activate the building of DEFINE CLUSTER and DELETE CLUSTER commands and extend existing data sets or create new data sets as required in response to an out of space condition.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 31% of the total text.

Page 1 of 4

Data Space Manager

This article describes a data space manager (DSM), which is a facility for controlling the operation of a computing system to define, delete, and extend data sets, thus permitting a user to concentrate on definitions of tables (constructs pertinent to the end user's needs) without being distracted by concerns about the characteristics of the direct access storage device (DASD) space in which the tables will reside CREATE and DROP TABLE commands are enhanced to automatically issue CREATE or DROP TABLESPACE commands which in turn activate the building of DEFINE CLUSTER and DELETE CLUSTER commands and extend existing data sets or create new data sets as required in response to an out of space condition.

One illustrative embodiment of the DSM facility may be implemented in the environment of the IBM Information Management System (IMS/VS) to insulate users from the job control language (JCL) statements needed to invoke OS/VS allocation and Access Method Services (AMS) and Virtual Sequential Access Method (VSAM) functions. After designating a list of available storage volumes in a "storage group", the user concentrates on communicating with a relational data base type component using the terminology of tables, indexes, and tablespaces. Datasets, a lower level concept, are then automatically created, expanded, reset, or dropped by DSM, as needed.

In the typical relational data base, the user deals with data in tabulated form where the tables can be pre-defined or dynamically constructed to meet the information needs of the user. The user, in general, has no interest in (and is likely to have an aversion to) knowing that the table definitions and the data that he accesses are physically stored in VSAM-like data sets on a particular set of DASD volumes.

The tables the user views reside in some "data base", and space within the data base is subdivided into separate "tablespaces". Tablespaces are defined explicitly by data base administrators (DBAs) Indexes reside in "indexspaces" within the data base. Indexspaces are defined implicitly as part of the process of defining an index. DSM must deal with both tablespaces and indexspaces, but by the time it receives control these have already been mapped into yet another level of logical storage, the "pageset space". The term "tablespace" will be used generically in the rest of this article to refer to tablespace, indexspace, and pageset space.

In addition to defining tablespaces on IMS subsystems DBAs are responsible for allocating the actual physical s ace needed DASD storage to "back" the logical space represented by each tablespace Allocation of the Physical space is done by creating VSAM Entry Sequenced Data Sets using the AMS service program to perform VSAM-related functions. For each data set needed in the IMS subsystem, the following AMS-defined command must be typed at a TSO terminal or entered as part of an OS/VS JCL jobstream: DEFINE CLUSTER (NAME (clustername) + ATTE...