Dynamic multicast routing facility for a distributed computing environment

• US 6,507,863 B2
• Filed: 01/27/1999
• Issued: 01/14/2003
• Est. Priority Date: 01/27/1999
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Novaes, Marcos N.
• Primary Examiner(s): Follansbee, John A.
• Application Number:
  US09/238,202
• Publication Number:
  US 20020165977A1
• Time in Patent Office: 1,448 Days
• Field of Search: 709/201, 709/227, 709/245
• US Class Current: 709/201
• CPC Class Codes: H04L 12/1877 Measures taken prior to transmission
• CPC Class Codes: H04L 12/4633 Interconnection of networks using encapsulation techniques, e.g. tunneling
• CPC Class Codes: H04L 45/16 Multipoint routing
• CPC Class Codes: H04L 45/46 Cluster building

Abstract

A Dynamic Multicast Routing (DMR) facility is provided for a distributed computing environment having a plurality of networks of computing nodes. The DMR facility automatically creates virtual interfaces between selected computing nodes of the networks to ensure multicast message reachability to all functional computing nodes within the distributed computing environment. The DMR facility employs a group of group leader nodes (GL_group) among which virtual interfaces for multicast messaging are established. Upon failure of one of the group leader nodes, another computing node of the respective network having the failing group leader node is assigned group leader status for re-establishing virtual interfaces. Virtual interfaces are established between the group leader nodes such that redundancy in message routing is avoided.

Claims

1. A dynamic method for ensuring multicast messaging within a distributed computing environment comprising:

establishing multiple groups of computing nodes within the distributed computing environment, wherein at least some of said multiple groups are located on heterogeneous networks that support multiple protocols;

selecting one node of each group of computing nodes as a group leader node;

forming a group of group leader nodes (GL_group) and selecting a group leader of the GL_group;

automatically creating a virtual interface for multicast messaging between the group leader node of the GL_group and at least one other group leader node within said GL_group, thereby providing protocol tunneling between groups of nodes of said distributed computing environment; and

wherein said automatically creating comprises for each group leader within said GL_group, determining whether said group leader is said GL_group leader, and if so, building a configuration file for tunneling from communication interfaces of said GL_group leader node to at least one other group leader in said GL_group, and performing protocol router daemon processing to build virtual interfaces between said GL_group leader and said at least one other group leader in said GL_group.

2. The dynamic method of claim 1, wherein said automatically creating comprises automatically creating a tunnel from the group leader node of the GL_group to each other group leader node within said GL_group, thereby ensuring multicast message routing between each group of said multiple groups of computing nodes of said distributed computing environment.

3. The dynamic method of claim 1, wherein said establishing multiple groups of computing nodes comprises for each computing node and for each communication interface of said computing node, determining a group identifier (groupID) from at least one of an IP address and a subnet mask for said communication interface.

4. The dynamic method of claim 3, wherein said determining said groupID comprises for each communication interface logically ANDing the IP address and the subnet mask to obtain a network identifier, and using said network identifier as a group identifier in grouping said computing nodes within the distributed computing environment.

5. The dynamic method of claim 1, wherein said selecting one node of each group of computing nodes as said group leader node comprises obtaining a membership list of all member nodes to said group of computing nodes and selecting as said group leader node a first successfully joined computing node to said group from said membership list.

6. The dynamic method of claim 1, wherein said selecting said group leader of said GL_group comprises obtaining a membership list of said GL_group and selecting as said group leader of said GL_group a first successfully joined group leader to said GL_group from said membership list.

7. The dynamic method of claim 1, wherein said automatically creating comprises for each group leader within said GL_group, determining whether said group leader is said GL_group leader, and if so, building a configuration file for mrouted tunneling from communication interfaces of said GL_group leader node to other group leaders in said GL_group, and performing mrouted daemon processing to build virtual interfaces between said GL_group leader and said other group leaders in said GL_group.

8. The dynamic method of claim 7, wherein if said group leader within said GL_group is other than said GL_group leader, said automatically creating comprises building a configuration file for mrouted tunneling from said group leader to said GL_group leader, and performing mrouted daemon processing to establish said virtual interface between said group leader and said GL_group leader.

9. The dynamic method of claim 1, further comprising automatically responding to a failure at at least one computing node of at least one group of computing nodes of the distributed computing environment by repeating said establishing, selecting, forming and automatically creating to ensure reachability of multicast messages to all functional computing nodes of said distributed computing environment.

10. The dynamic method of claim 1, wherein said automatically creating comprises creating said virtual interface for multicast messaging between the group leader of the GL_group and at least one other group leader node within the GL_group such that redundancy in routing of multicast messages is avoided within the distributed computing environment.

11. A system for ensuring IP multicast messaging within a distributed computing environment comprising:

multiple groups of computing nodes within the distributed computing environment, wherein at least some of said multiple groups are located on heterogeneous networks that support multiple protocols;

means for selecting one node of each group of computing nodes as a group leader node;

means for forming a group of group leader nodes (GL_group) and selecting a group leader of the GL_group;

means for automatically creating a virtual interface for multicast messaging between the group leader node of the GL_group and at least one other group leader node within said GL_group, thereby providing protocol tunneling between groups of nodes of the distributed computing environment; and

wherein said means for automatically creating comprises for each group leader within said GL_group, means for determining whether said group leader is said GL_group leader, and if so, for building a configuration file for tunneling from communication interfaces of said GL_group leader node to at least one other group leader in said GL_group, and for performing protocol router daemon processing to build virtual interfaces between said GL_group leader and said at least one other group leader in said GL_group.

12. The system of claim 11, wherein said means for automatically creating comprises-means for automatically creating a tunnel from the group leader node of the GL_group to each other group leader node within said GL_group, thereby ensuring multicast message routing between each group of said multiple groups of computing nodes of said distributed computing environment.

13. The system of claim 11, wherein said means for establishing multiple groups of computing nodes comprises means disposed at each computing node for determining for each communication interface of said node a group identifier (groupID) from at least one of an IP address and a subnet mask for said communication interface.

14. The system of claim 13, wherein said means for determining said groupID comprises means for each communication interface for logically ANDing the IP address and the subnet mask to obtain a network identifier, and for using said network identifier as a group identifier in grouping said computing nodes within said distributed computing environment.

15. The system of claim 11, wherein said means for selecting one node of each group of computing nodes as said group leader node comprises means for obtaining a membership list of all member nodes to said group of computing nodes and for selecting as said group leader node a first successfully joined computing node to said group from said membership list.

16. The system of claim 11, wherein said means for selecting said group leader of said GL_group comprises means for obtaining a membership list of said GL_group and for selecting as said group leader of said GL_group a first successfully joined group leader to said GL_group from said membership list.

17. The system of claim 11, wherein said means for automatically creating comprises for each group leader within said GL_group, means for determining whether said group leader is said GL_group leader, and if so, for building a configuration file for mrouted tunneling from communication interfaces of said GL_group leader node to other group leaders in said GL_group, and means for performing mrouted daemon processing to build virtual interfaces between said GL_group leader and said other group leaders in said GL_group.

18. The system of claim 11, further comprising means for automatically responding to a failure at at least one computing node of at least one group of computing nodes of the distributed computing environment by repeating said means for establishing, means for selecting, means for forming, and means for automatically creating to ensure reachability of multicast messages to all functional computing nodes of the distributed computing environment.

19. The system of claim 11, wherein said means for automatically creating comprises means for creating said virtual interface for multicast messaging between said group leader of the GL_group and at least one other group leader node within the GL_group such that redundancy in routing of multicast messages is avoided within the distributed computing environment.

20. An article of manufacture comprising:

a computer program product comprising a computer usable medium having computer readable program code means therein for use in providing multicast messaging within a distributed computing environment, said computer readable program code means in said computer program product comprising;

(i) computer readable program code means for causing a computer to effect establishing multiple groups of computing nodes within the distributed computing environment, wherein at least some of said multiple groups are located on heterogeneous networks that support multiple protocols;

(ii) computer readable program code means for causing a computer to effect selecting one node of each group of computing nodes as a group leader node;

(iii) computer readable program code means for causing a computer to effect forming a group of group leader nodes (GL_group) and selecting a group leader of the GL_group;

(iv) computer readable program code means for causing a computer to effect automatically creating a virtual interface for IP multicast messaging between the group leader-node of the GL_group and at least one other group leader node within said GL_group, thereby providing protocol tunneling between groups of nodes of said distributed computing environment; and

wherein said automatically creating comprises for each group leader within said GL_group, determining whether said group leader is said GL_group leader, and if so, building a configuration file for tunneling from communication interfaces of said GL_group leader node to at least one other group leader in said GL_group, and performing protocol router daemon processing to build virtual interfaces between said GL_group leader and said at least one other group leader in said GL_group.

21. The article of manufacture of claim 20, wherein said computer readable program code means for causing a computer to effect automatically creating comprises computer readable program code means for causing a computer to effect automatically creating a tunnel from the group leader node of the GL_group to each other group leader node within the GL_group, thereby ensuring multicast message routing between each group of said multiple groups of computing nodes of the distributed computing environment.

22. The article of manufacture of claim 20, wherein said computer readable program code means for causing a computer to effect automatically creating comprises computer readable program code means for causing a computer to effect creating said virtual interface for multicast messaging between the group leader of the GL_group and at least one other group leader node within the GL_group such that redundancy in routing of multicast messages is avoided within the distributed computing environment.

23. The article of manufacture of claim 20, further comprising computer readable program code means for causing a computer to effect automatically responding to a failure at at least one computing node of at least one group of computing nodes of the distributed computing environment to ensure reachability of multicast messages to all functional computing nodes of said distributed computing environment.

24. A system for ensuring multicast messaging within a distributed computing environment comprising:

multiple groups of computing nodes within the distributed computing environment, wherein at least some of said multiple groups are located on heterogeneous networks that support multiple protocols;

a processor associated with the distributed computing environment; and

code executable by said processor associated with said distributed computing environment, said code causing said processor to effect;

selecting one node of each group of computing nodes as a group leader node;

forming a group of group leader nodes (GL_group) and selecting a group leader of the GL_group;

automatically creating a virtual interface for multicast messaging between the group leader node of the GL_group and at least one other group leader node within said GL_group, thereby providing protocol tunneling between groups of nodes of said distributed computing environment; and

wherein said automatically creating comprises for each group leader within said GL_group, determining whether said group leader is said GL_group leader, and if so, building a configuration file for tunneling from communication interfaces of said GL_group leader node to at least one other group leader in said GL_group, and performing protocol router daemon processing to build virtual interfaces between said GL_group leader and said at least one other group leader in said GL_group.