Browse Prior Art Database

Dynamic Memory for Program Staging

IP.com Disclosure Number: IPCOM000074389D
Original Publication Date: 1971-Apr-01
Included in the Prior Art Database: 2005-Feb-23
Document File: 3 page(s) / 59K

Publishing Venue

IBM

Related People

Earle, JG: AUTHOR [+3]

Abstract

This system provides hardware assisted program staging (I/O initiated rollin/rollout) to a bulk memory subsystem. The basic system configuration is seen in drawing A. Central processing unit 1 is connected via bus 3 to main storage 5. The main storage is also connected via buses 7 and 7a to bulk storage unit 9 by way of staging and scheduler subsystem 11, for moving programs between main storage and bulk storage. Bulk storage unit 9 is also connected over bus 13 by way of staging and scheduler subsystem 11 to the system I/O devices. Line 15 is connected from the CPU to the staging and scheduler subsystem and, when active, indicates that an I/O instruction has been issued by the program in execution.

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 54% of the total text.

Page 1 of 3

Dynamic Memory for Program Staging

This system provides hardware assisted program staging (I/O initiated rollin/rollout) to a bulk memory subsystem. The basic system configuration is seen in drawing A. Central processing unit 1 is connected via bus 3 to main storage 5. The main storage is also connected via buses 7 and 7a to bulk storage unit 9 by way of staging and scheduler subsystem 11, for moving programs between main storage and bulk storage. Bulk storage unit 9 is also connected over bus 13 by way of staging and scheduler subsystem 11 to the system I/O devices. Line 15 is connected from the CPU to the staging and scheduler subsystem and, when active, indicates that an I/O instruction has been issued by the program in execution. In essence, when a program in execution in the CPU causes an I/O instruction to be issued, the entire program is moved out of main storage by way of bus 7 and relocated in the bulk storage unit by way of staging and scheduler subsystem 11. The I/O instruction is issued, such that the data transfer occurs between the I/O device and the bulk storage unit. On termination of the transfer, the program is returned to main storage and re- entered for execution from the operating system. Thus, the entire user program's main storage space is available to other users while his I/O is being performed.

The algorithm incorporated into the system is as follows:
1) Move tasks from main storage to bulk storage when they

request I/O, but before they go into the wait stage.
2) Bring in other available tasks from bulk storage via the

scheduler to main storage for execution when main storage

is available.
3) Execute the I/O sequence from the bulk device to the I/O

device, accepting or sending data from the task's location

in the bulk storage device.
4) After the I/O operation is complete, mark the task as

available for returning to main storage.

A simplified flow chart for the operation of the system for a normal I/O operation, is seen in drawing B. As seen at 21, all normal programming results in GET/PUT macros in the calling program. This eventually causes the issuing of the start I/O instruction. At 23, the I/O supervisor sets up the I/O channel program in various...