Browse Prior Art Database

Clock Source Selection Method in Distributed Communication System

IP.com Disclosure Number: IPCOM000045894D
Original Publication Date: 1983-Apr-01
Included in the Prior Art Database: 2005-Feb-07
Document File: 6 page(s) / 42K

Publishing Venue

IBM

Related People

Rogers, LS: AUTHOR

Abstract

Synchronous communication systems have a need for clocking throughout the system For simplicity's sake, the system is described as consisting of a number of nodes interconnected in network fashion with each node connected to one or more other nodes via bidirectional digital communication links. Ordinarily, clocking information is shared throughout such a system either via a particular scheme of recovering clock information from data transitions or of encoding the data in such a way as to make clock recovery independent of the data transitions. These specific techniques are well known and are not discussed further. The present article addresses the problem of selecting which node or element in the network will be selected for providing the clock information for entire system synchronization.

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 19% of the total text.

Page 1 of 6

Clock Source Selection Method in Distributed Communication System

Synchronous communication systems have a need for clocking throughout the system For simplicity's sake, the system is described as consisting of a number of nodes interconnected in network fashion with each node connected to one or more other nodes via bidirectional digital communication links. Ordinarily, clocking information is shared throughout such a system either via a particular scheme of recovering clock information from data transitions or of encoding the data in such a way as to make clock recovery independent of the data transitions. These specific techniques are well known and are not discussed further. The present article addresses the problem of selecting which node or element in the network will be selected for providing the clock information for entire system synchronization. The difficulty is compounded since two or more streams of potential data and/or clocking information may be received at a given node from two or more other nodes in the same system. Additionally, the , configuration of the system may be changed by the failure of a node or the insertion of additional nodes.

The present article shows a means of providing synchronization to automatically adapt for failures or configuration changes.

The figure illustrates a generalized form of a distributed communication system comprising four nodes A, B, C and D, as shown. Node D, for example, may arbitrarily be connected to some primary digital communications facility or even a host computer. The choice of node D for the connection is completely arbitrary. A secondary digital communication facility is shown attached to node C. This could be a multi-plexer or communication controller of some sort, or another computer. Each of the nodes may comprise one or more computers and/or input/output terminals or any other variety of communications equipment.

In the figure, the assumption is made that each node has sufficient programming and execution capability to contain a program in either software or hardware that will make each node equally capable of serving as a system clock master and which will embody sufficient logic to enable the scheme to be presented to be conducted. A system clock supply configuration is established through a technique of bidding carried on through the communications paths among the various nodes. The bidding sessions are to be carried on at specified intervals, such as turn-on of the system or when any change in system configuration or operation occurs. Any time a new node is brought into service, a new bidding session will be established. Any time a node detects a failure or a neighbor, a new bidding session may be requested. These dynamic configuration changes and the system' s methodical adaptation thereto accommodate current operational configurations.

In addition, communications systems such as that depicted in the figure often connect to other systems through common carrie...