Distributed control of alias name usage in networks

• US 4,718,005 A
• Filed: 05/03/1984
• Issued: 01/05/1988
• Est. Priority Date: 05/03/1984
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Feigenbaum, Barry A., Gibbs, Dennis D., Sachsenmaier, Robert
• Primary Examiner(s): Harkcom, Gary V.
• Assistant Examiner(s): CLARK, DAVID
• Application Number:
  US06/604,684
• Time in Patent Office: 1,342 Days
• Field of Search: 340/825, 340/825.05, 340/825.07, 340/825.52, 340/825.03, 370/85, 370/94, 364/200 MS File, 364/900 MS File
• US Class Current: 709/222
• CPC Class Codes: H04L 2101/365 Application layer names, e.g. buddy names, unstructured names chosen by a user or home appliance name
• CPC Class Codes: H04L 61/3005 Mechanisms for avoiding name conflicts
• CPC Class Codes: H04L 61/3025 Domain name generation or assignment
• CPC Class Codes: H04L 61/45 Network directories; Name-to-address mapping
• CPC Class Codes: H04L 69/00 Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass networks security protocols H04L9/40; wireless communication networks H04W

Abstract

Computer systems linked to nodes in a communication network communicate directly with each other to establish name associations for entities (e.g. programs, storage files, etc.) susceptible of being shared across the network, and thereafter communicate further by names (rather than addresses) to establish sessions (i.e. logical connections over the network) between entities. Each system maintains a table of names assigned to locally accessible entities. The name and session establishment processes are architected to avoid dependence on any "central" or "master" system, and to allow for extension of communications relating to these processes across devices linking portions of the network associated with different communication media or channels.

Claims

1. For a data communication network comprising multiple computing nodes linked by communication media and/or channels, each node serving one or more entities, a method of establishing logical name associations between nodes and their respectively served entities, for permitting programs and/or operators at said nodes to share entities dispersed across said network and initiate communications with said entities by associated names without having to determine physical locations of said entities, comprising:

maintaining name association tables at said nodes indicating names adopted at said nodes for respectively served entities;

at any node seeking to adopt a new name association, broadcasting a Name Check request message over said network indicating the name to be adopted and the physical address location in said network of the node proposing the adoption;

receiving said request at other nodes, and comparing the proposed name with names stored in the name association tables at said other nodes;

at any of said other nodes finding a name in their table matching the proposed name, transmitting an acknowledgment message over said network addressed specifically to the node which originated said Name Check request; and

at the node which originated said Name Check, adopting or rejecting said name conditional upon the number of said acknowledgments received in response to said Name Check.

2. A method of establishing name associations as defined in claim 1 wherein said adoption or rejection of the proposed name, by the node which originated said Name Check request, is determined by:

timing out a predetermined time interval after transmission of said Name Check request;

counting the number of acknowledgments received during said interval;

rejecting the name if one or more acknowledgments are received during said interval; and

adopting the name if no acknowledgments are received during said interval.

3. A method of establishing name associations as defined in claim 1 comprising:

repeating said Name Check broadcasting and comparing and acknowledging actions for a predetermined number of times;

timing out a predetermined interval at the node originating said Name Check after each broadcasting of said Name Check and before the next broadcasting thereof;

in each said interval, monitoring the acknowledgments received from other nodes in response to the respective Name Check broadcast;

adopting the proposed name if no acknowledgments are received during any one of said timeout intervals; and

rejecting the proposed name if one or more acknowledgments are received during each of said timeout intervals.

4. The name adoption method of either claim 2 or claim 3 comprising:

broadcasting a Duplicate Use message over said network if more than one acknowledgement is received during a said timeout interval at the node which broadcast said Name Check request message, said Duplicate Use message indicating said proposed name and its adoption at more than one of said acknowledging nodes;

receiving said Duplicate Use message at said other nodes which received said Name Check message, and comparing said proposed name with names stored in respective said name association tables;

at said other nodes having matching name entires in respective name tables, entering into respective tables Duplicate Use indications in association with the respective matched name entries, each said indication denoting use of the associated name at more than one of said nodes; and

at said other nodes having Duplicate Use indications entered into respective name tables, inhibiting operations requiring use of the respective matching name.

5. The method of name adoption of claim 1 comprising:

at any node seeking to adopt a new name association, checking the respective name table for entries indicating existing association of the name with an entity;

rejecting adoption of the new name association if there is an existing local association, and in that instance eliminating said Name Check broadcasting step; and

proceeding with said Name Check broadcasting step and other operations conditional thereon only if a local association is not found.

6. For a data communication network comprising multiple computing nodes, each node serving one or more entities which may be shared by other nodes, each node having a physical address in said network and being capable of having one or more alias names associated with either the node or its respectively served entities, a method for establishing a communication session between a pair of entities having associated alias names, using only the associated names and not information regarding the physical locations of the entities, comprising:

maintaining name association tables at said nodes for storing indications of logical names associated with respectively served entities;

at a node seeking to establish a session between a respectively served first entity having a first name and a second entity having a second name, comparing said second name to the names stored in the name association table maintained at the respective node;

establishing a session between said first and second entities, by means of processing operations performed entirely within the respective node--i.e. without any communications over said network--if the second name matches a name stored in the respective name table; and

if the second name does not match any name in the respective name table, successively

broadcasting a Call Name check message over said network from the node seeking to establish said session, said message containing said second name and the address in said network of the node originating the broadcast;

timing out a predetermined time interval at said originating node immediately following said broadcast;

receiving said Call Name check message at other nodes in said network;

at each said other node, comparing the second name in said received message with names stored in the respective table and transmitting an acknowledgment message directed by address specifically to the node which broadcast the Name Check message if and only if a match is detected between said second name and a name in the respective table;

at the node which originated the Name Check broadcast, monitoring acknowledgments of said broadcast received during said timeout interval; and

conditional upon said received acknowledgments, either initiating further communications with an acknowledging other node, for continuing the process of seeking to establish said session, or aborting said process.

7. The method for establishing a communication session of claim 6 wherein said further communications include:

sending a call request communication from said node seeking to continue said process to said second entity; and

returning an acknowledgement in response to said call request from the node serving said second entity to the node which sent the call request.

8. The method of claim 7 wherein said nodes are adapted for establishing Listening State associations between respectively served entities and other entities, for regulating communication traffic between entities, and said second acknowledgment is given only on condition that a Listening State association pre-exists between said second name and said first name at a node serving an entity associated with said second name and receiving a said Call Request.

9. The session method of claim 6 including:

continuing said process if only one acknowledgement is received in response to said Name Check broadcast and aborting said process if either no acknowledgement or more than one acknowledgement is received in response to said Name Check broadcast; and

in response to receipt of more than one acknowledgement of said Name Check broadcast, broadcasting a Duplicate Use message over said network from said node which originated said Name Check broadcast, said Duplicate Use message indicating existance of conflicting name usage and having the effect of informing the computing systems at the nodes which acknowledged the Name Check broadcast that their uses of said second name are conflicting.