Browse Prior Art Database

IMPL Lockout During Disk Enclosure Save/ Restore

IP.com Disclosure Number: IPCOM000049214D
Original Publication Date: 1982-May-01
Included in the Prior Art Database: 2005-Feb-09
Document File: 2 page(s) / 14K

Publishing Venue

IBM

Related People

Koeller, PD: AUTHOR [+2]

Abstract

In a computer system which implements a single-level store concept, it is desirable to provide a way to save and restore the data from one of the disk enclosures without requiring the entire single-level store to be saved and restored. This function is useful when one of the disk enclosures, which make up the single level store, is being replaced because of a problem within the disk enclosure. Because normal operation of the system between the time the data on the disk enclosure has been saved and then restored will result in mismatches in the single level store directories. some method is needed to prevent normal system operation during this time.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

IMPL Lockout During Disk Enclosure Save/ Restore

In a computer system which implements a single-level store concept, it is desirable to provide a way to save and restore the data from one of the disk enclosures without requiring the entire single-level store to be saved and restored. This function is useful when one of the disk enclosures, which make up the single level store, is being replaced because of a problem within the disk enclosure. Because normal operation of the system between the time the data on the disk enclosure has been saved and then restored will result in mismatches in the single level store directories. some method is needed to prevent normal system operation during this time.

The IBM System/38 is implemented using a virtual memory concept referred to as a single-level store. In this concept, the Main Storage and Disk Storage are treated as a single storage space. The concept uses directories to keep track of where various objects reside in the storage space.

In the event that one of the disk enclosures, which is part of the single-level store, becomes unusable and must be replaced, it is useful to have the data from the failing disk enclosure on tapes or diskettes. The failing disk enclosure may then be replaced and the data restored to a new disk enclosure.

Since normal operation of the system continually updates the single level store directories, some way must be provided to run the system in a maintenance mode which does not change the contents of disk storage or modify the single- level store directories. This is accomplished by loading the necessary programs from diskette and executing them in Main Storage without the use of the disk storage.

One of the programs which is provided in this maintenance mode, called the save program, reads the contents of a disk enclosure and writes the data on tapes or diskettes. Another such program, called the restore program, reads the data from the tapes or diskettes and writes the data to a disk enclosure.

Between the time that the data is saved on tapes or diskettes and then restored to the new disk enclosure, it is necessary to prevent normal operation of the system. This is necessary to prevent a mismatch in the single-level store directories which are used to indicate where all objects reside in the single-level store. Several possible solutions to this problem were presented and evaluated.

The first solution which was recommended was to have the operator leave a handwritten note on the system which warns that the system should not be run in normal mode. This solution was rejected because it involved a manual operation which the operator may forget to do. The second solution suggested was to have the save program display the warning message. This solution was rejected because the system must be powered off to remove the disk enclosure, and thus the warning message would be lost.

The actual solution which was implemented involves the use of two levels of ...