Browse Prior Art Database

Controlling Error Propagation in a Mass Storage System Serving A Plurality of CPU Hosts

IP.com Disclosure Number: IPCOM000088194D
Original Publication Date: 1977-May-01
Included in the Prior Art Database: 2005-Mar-04
Document File: 2 page(s) / 48K

Publishing Venue

IBM

Related People

Hoff, CE: AUTHOR [+4]

Abstract

A mass storage system (MSS) services a plurality of central processing unit (CPU) hosts via a plurality of independent I/O channels as well as via a control link to a mass storage control (MSC) of MSS. Each host has a plurality of independent I/0 channels to one or more staging adaptors (INSA) within a staging storage facility (SSF). One or more of the INSAs are connected to a Plurality of direct-access storage device (DASD) spindles which are independently controlled by their respective INSAs.

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 2

Controlling Error Propagation in a Mass Storage System Serving A Plurality of CPU Hosts

A mass storage system (MSS) services a plurality of central processing unit (CPU) hosts via a plurality of independent I/O channels as well as via a control link to a mass storage control (MSC) of MSS. Each host has a plurality of independent I/0 channels to one or more staging adaptors (INSA) within a staging storage facility (SSF). One or more of the INSAs are connected to a Plurality of direct-access storage device (DASD) spindles which are independently controlled by their respective INSAs.

Under control of MSC, data is staged and destaged to and from SSF from and to a data base storage facility MSF. MSC controls both the SSF and the MSF. CPU hosts operate using virtual address space for accessing data in SSF. During the course of normal operations a CPU host, through its computer programming, can detect an error in the supplied data; i.e., the data supplied over the I/O channel is in error.

In response to such error, rather than abnormally ending (ABEND) the job at hand, the CPU host retries the operation to that portion of the virtual address space called a virtual page. In this regard, a virtual page is a logical address space forming a portion of a virtual volume (VV). A VV is a virtually mounted DASD pack address space which can be assigned to a virtual unit address (VUA) (virtually addressed unit) within SSF. Such VV can be logically mounted on a VUA. The VUA can be assigned address space on the real DASD spindles in accordance with the predistribution algorithm for maximizing data availabi...