Browse Prior Art Database

Computer System Run Time Structure

IP.com Disclosure Number: IPCOM000081784D
Original Publication Date: 1974-Aug-01
Included in the Prior Art Database: 2005-Feb-28
Document File: 3 page(s) / 43K

Publishing Venue

IBM

Related People

Bains, RL: AUTHOR [+4]

Abstract

In a multiprogramming computer system operating in a data base environment, it is desirable to (1) make programs independent of the data they use so that program logic may be shared among several users; and (2) make the data independent of user programs so that a piece of data may be shared by several users. Contemporary machine architectures have either not provided for (1), (2) or provide (1) and (2) in limited form only for the "data base" data; that data which is subject to I/O reads and writes. Such provisions were in low-performance software written atop the machine interface.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Computer System Run Time Structure

In a multiprogramming computer system operating in a data base environment, it is desirable to (1) make programs independent of the data they use so that program logic may be shared among several users; and (2) make the data independent of user programs so that a piece of data may be shared by several users. Contemporary machine architectures have either not provided for
(1), (2) or provide (1) and (2) in limited form only for the "data base" data; that data which is subject to I/O reads and writes. Such provisions were in low- performance software written atop the machine interface.

Fig. 1 shows a high-level system (HLS) structure which provides a fresh approach to the problem of decoupling program logic and data. This HLS consists of: (A) Instruction stream (code). (B) Data descriptor table (DDT). (C) Data. Instruction stream (A) consists of: Operator references. Operand references.

Operator references are encoded values which could represent an index into a predefined table of system operators. Operand references are encoded values representing an index into a program execution-dependent able of operand data descriptors. The instruction stream operand reference contains no information about the operand. The only dependency of the instruction stream on the data descriptor table (DDT) is an ordering dependency; descriptor X must continue to represent the same object throughout the life of a program, if the program logic is to execute correctly.

DDT (B) consists of: Data Attributes. - type - length and scaling - optional name - etc.

Access Control Indicators (see Fig. 2). - read-only indicator - security lock indicator - monitor indicator - accounting required - etc.

Data Value Linkage.

The DDT can be used in the following ways: The DDT is free to change independent of program logic (instruction stream). Any attribute of any data item may be modified without modifying the program. This feature allows a data base system to "fill in" the descriptor entries at "open time" and, therefore, provide very late binding of instruction stream and data base data.

Multiple copies of the DDT may be used with a single copy of the instruction stream.

A single DDT...