Browse Prior Art Database

Duplexed Local Communications Controllers

IP.com Disclosure Number: IPCOM000043584D
Original Publication Date: 1984-Sep-01
Included in the Prior Art Database: 2005-Feb-05
Document File: 3 page(s) / 35K

Publishing Venue

IBM

Related People

DeVoe, MP: AUTHOR [+2]

Abstract

This article describes a method to provide fault tolerance in a multiple processor IBM Series/1 configuration by supporting two local communications controller rings and being able to switch back and forth between them. The Series/1 is a 16-bit minicomputer that supports standard data processing, realtime, and transaction-oriented applications. It supports the attachment of one or more local communications controllers which permit multiple Series/1 processors to be joined into a single ring of machines where each machine is able to communicate with all of the others at relatively high speeds. The local communications controller (LCC), which is also known as the ring, is the hardware medium that ties together processors in a multiple processor Series/1 configuration.

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 52% of the total text.

Page 1 of 3

Duplexed Local Communications Controllers

This article describes a method to provide fault tolerance in a multiple processor IBM Series/1 configuration by supporting two local communications controller rings and being able to switch back and forth between them. The Series/1 is a 16-bit minicomputer that supports standard data processing, realtime, and transaction-oriented applications. It supports the attachment of one or more local communications controllers which permit multiple Series/1 processors to be joined into a single ring of machines where each machine is able to communicate with all of the others at relatively high speeds. The local communications controller (LCC), which is also known as the ring, is the hardware medium that ties together processors in a multiple processor Series/1 configuration. In order to provide fault tolerance, the system must be able to support two rings and be able to switch back and forth between them. In a fault-tolerant system composed of multiple minicomputers such as the IBM Series/1, the medium that ties together the individual machines into a single entity is vital to the continued operation of the system. Should this medium fail, the individual machines are isolated from each other and the system degrades or is totally lost. Thus, if the system is to be fault-tolerant, the system must have and be able to use some alternative communications medium to continue the operation of the system. To solve the fault-tolerance problem the machines in the system can be attached to a second ring. This ring serves as a backup ring if the first ring, which is used for normal inter-machine message traffic, fails. One of the major problems when a failure occurs in a multiple processor system is how to determine what has gone wrong and where. In the duplexed ring system described herein, there is a system task that uses the backup ring to monitor the status of all of the processors on the ring. Every two minutes a designated node on the ring writes on the backup ring to all of the other nodes. When each node receives this write, it writes back to...