Multiprocessor cluster membership manager framework

• US 5,666,486 A
• Filed: 06/23/1995
• Issued: 09/09/1997
• Est. Priority Date: 06/23/1995
• Status: Expired due to Term

• Current Assignee: Data General Corp. (Dell Technologies Inc.)
• Original Assignee: Data General Corp. (Dell Technologies Inc.)
• Inventors: Compton, James T., McGrath, Paul T., Vook, Eric R., Soufi, Khaled S., Thorstad, Brian J., Huber, Andrew R., Alfieri, Robert A.
• Primary Examiner(s): Olms, Douglas W.
• Assistant Examiner(s): PHILLIPS, MATTHEW
• Application Number:
  US08/493,550
• Time in Patent Office: 809 Days
• Field of Search: 395/200.01, 395/200.02, 395/200.03, 395/200.04, 395/200.05, 395/200.06, 395/200.08, 395/200.1, 395/200.12, 395/650, 364/131, 364/132, 364/133, 364/134, 364/942.3, 364/942.4, 364/942.5, 364/942.51, 364/942.6
• US Class Current: 709/217
• CPC Class Codes: G06F 11/1425 by reconfiguration of node membership
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements

Abstract

A shared-disk cluster system includes a cluster membership manager framework which coordinates the joining or leaving among all nodes in a cluster including taking the multiple layers of involved subsystems through transitions. Subsystems are notified of transitions in particular order depending upon the transition, and all nodes'"'"' subsystems receiving a notification must process that notification prior to another layer of subsystems being notified. One of the subsystems registered for notification is an event manager in user space. The event manager carries out transfers of client services, including user applications, resulting from nodes joining and leaving the cluster. This includes a registration and launch service which registers a node, or multiple nodes, in a cluster which claims, or is assigned, responsibility for the service and provides an optional launching function which initiates the client service upon successful registration.

Claims

1. In a multiprocessor system having multiple nodes, a shared resource accessible to all nodes and multiple interdependent levels of subsystems on each of said nodes, a method of combining particular ones of said nodes in a cluster so that it appears to users that the entire cluster is substantially a single node containing said shared resource, said method including notifying all levels of subsystems running on nodes presently in the cluster of transitions of nodes joining and leaving the cluster in order to provide a consistent view of active membership in the cluster.

2. The method of claim 1 wherein the subsystems are interdependent in levels, with higher level subsystems dependent on the operation of lower level subsystems.

3. The method of claim 2 including notifying one of said subsystems on all of said nodes presently in the cluster of a transition, processing that notification at said one of said subsystems prior to notifying another of said subsystems on all of said nodes presently in the cluster of the transition.

4. The method of claim 3 including notifying subsystems beginning with lower level subsystems and proceeding in sequence through higher levels of subsystems of a transition of a node joining the cluster.

5. The method of claim 3 including notifying subsystems beginning with higher lever subsystems and proceeding in sequence through lower levels of subsystems of a transition of a node gracefully leaving the cluster.

6. The method of claim 3 including notifying subsystems beginning with lower level subsystems and proceeding in sequence through higher levels of subsystems of a transition of a node forced from the cluster by other processors.

7. The method of claim 2 wherein said subsystems include a higher level subsystem which interacts with user programs.

8. The method of claim 7 wherein said higher level subsystem includes a service which automatically and atomically transfers user programs to other nodes when the node executing the user programs leaves the cluster.

9. The method of claim 2 wherein said subsystems include a distributed lock manager subsystem, a virtual disk manager subsystem and a shared file subsystem.

10. The method of claim 1 wherein for a transition of a node joining the cluster, said method includes the steps of:

a) registering subsystems of the joining node to receive transition notifications;

b) joining the node to the cluster; and

c) notifying registered subsystems in the cluster that the joining node has joined the cluster.

11. The method of claim 1 wherein for a transition of one node being forced out of the cluster by another node, said method includes the steps of:

a) the another node notifying registered subsystems that the one node is being forced out of the cluster; and

b) transferring registered programs executing on said one node atomically to a different node and recovering the programs to execute on said different node.

12. In a multiprocessor system having multiple nodes, a shared resource accessible to all nodes and multiple interdependent levels of subsystems on each of said nodes, a method of initiating client services on particular ones of said nodes in a cluster so that it appears to said client services that the entire cluster is substantially a single node containing said shared resource, said method including choosing a node for each client service, registering the client service with that node, and notifying nodes presently in the cluster that the particular client service is registered with the particular node, whereby the particular service can be transferred to another node if the particular node leaves the cluster.

13. The method of claim 12 further including launching a client service on a node according to an action parameter included with the client service in response to registering that client service with that node.

14. The method of claim 13 further including grouping client services as a parent client service and at least one child client service, registering grouped client services with the same node and launching grouped client services according to an action parameter included with the parent client service.

15. The method of claim 12 wherein said choosing a node includes providing a database of choosing factors for the client service and applying said choosing factors to the nodes presently in the cluster, said choosing factors establishing rules relating nodes to the client service.

16. The method of claim 15, wherein said choosing factors are selected from the group including allowable nodes, disallowable nodes and node preferences.

17. The method of claim 12 further including notifying nodes joining the cluster that the particular client service is registered with the particular nodes.

18. The method of claim 12 including monitoring a client service registered with a node at that node using an event watcher.

19. The method of claim 18 including enabling the event watcher in response to registering the client service and disabling the event watcher in response to de-registering the client service.

20. The method of claim 12 wherein said multiple subsystems includes a cluster membership manager which controls which of said nodes are presently in the cluster and wherein said cluster membership manager provides notification of which nodes are in the cluster.

21. The method of claim 20 wherein said initiating services on a particular one of said nodes is performed by another one of said multiple subsystems.

22. A multiprocessor cluster system having multiple nodes, a shared resource accessible to all nodes, a cluster communication medium between said nodes, and multiple subsystems on each of said nodes, comprising:

a cluster membership manager subsystem adapted to notify subsystems running on nodes presently in the cluster of transitions of nodes joining and leaving the cluster in order to provide a consistent view of active membership in the cluster;

an event manager subsystem adapted to detect and react to cluster errors; and

a registration and launch service responsive to said event manager and adapted to initiate client services on particular ones of said nodes in a cluster so that it appears to said client services that the entire cluster is substantially a single node containing said shared resource, wherein said registration and launch subsystem chooses a node for each client service, registers the client service with that node, and notifies nodes presently in the cluster that the particular service is registered with the particular node.

23. The multiprocessor cluster system in claim 22 wherein the subsystems are interdependent in levels, with higher level subsystems dependent on the operation of lower level subsystems.

24. The multiprocessor cluster system in claim 23 wherein said membership manager subsystem notifies one of said subsystems on all of said nodes presently in the cluster of a transition, and that one of said subsystems on all nodes processes that notification prior to said membership manager subsystem notifying another of said subsystems on all of said nodes presently in the cluster of the transition.

25. The multiprocessor cluster system in claim 24 wherein said membership manager notifies subsystems beginning with lower level subsystems and proceeding in sequence through higher levels of subsystems of a transition of a node joining the cluster.

26. The multiprocessor cluster system in claim 24 wherein said membership manager notifies subsystems beginning with higher level subsystems and proceeding in sequence through lower levels of subsystems of a transition of a node gracefully leaving the cluster.

27. The multiprocessor cluster system in claim 24 wherein said membership manager notifies subsystems beginning with lower level subsystems and proceeding in sequence through higher levels of subsystems of a transition of a node forced from the cluster by other processors.

28. The multiprocessor cluster system in claim 22 wherein said registration and launch service launches a client service on a node according to an action parameter included with the client service in response to registering that client service with that node.

29. The multiprocessor cluster system in claim 28 wherein said registration and launch service is adapted to group client services as a parent client service and at least one child client service, and wherein said registration and launch service further registers grouped client services with the same node and launches grouped client services according to an action parameter included with the parent client service.

30. The multiprocessor cluster system in claim 22 wherein said registration and launch service includes a database of choosing factors for the client service and applies said choosing factors to the nodes presently in the cluster to choose the node for registering a client service, said choosing factors establishing rules relating nodes to the client service.

31. The multiprocessor cluster system in claim 22 wherein said registration and launch service further notifies nodes joining the cluster that the particular client service is registered with the particular node.

32. The multiprocessor cluster system in claim 22 wherein said event manager includes an event watcher for monitoring a client service registered with a node at that node.

33. A computer usable medium in which program code is embodied, said program code defining an operating system for a multiprocessor cluster system having multiple nodes, a shared resource accessible to all processors, and including multiple interdependent levels of subsystems, one of said subsystems being a cluster membership manager subsystem which defines a computer process for combining particular ones of said nodes in a cluster so that it appears to users that the entire cluster is substantially a single node containing the shared resource, said cluster membership manager subsystem adapted to notify all levels of subsystems running on nodes presently in a cluster of transitions of nodes joining and leaving the cluster in order to provide a consistent view of active membership in the cluster.

34. A computer usable medium in which program code is embodied, said program code defining an operating system for a multiprocessor cluster system having multiple nodes, a shared resource accessible to all processors, and including a registration and launch service which defines a computer process for allocating said nodes to perform client services, said registration and launch service adapted to initiate client services on particular ones of nodes so that it appears to said client services that the entire cluster is substantially a single node containing said shared resource, wherein said registration and launch service chooses a node for each client service, registers the client service with that node, and notifies nodes presently in the cluster that the particular service is registered with the particular node.

35. A computer usable medium in which program code is embodied, said program code defining an operating system for a multiprocessor cluster system having multiple nodes, and a shared resource accessible to all nodes, comprising:

multiple subsystems that are interdependent in levels, with higher level subsystems dependent on the operation of lower level subsystems;

one of said subsystems comprising a cluster membership manager subsystem which defines a computer process for combining particular ones of said nodes in a cluster so that it appears to users that the entire cluster is substantially a single node containing the shared resource, said cluster membership manager subsystem adapted to notifying subsystems running on nodes presently in a cluster of transitions of processors joining and leaving the cluster in order to provide a consistent view of active membership in the cluster; and

one of said subsystems including a registration and launch service which defines a computer process for allocating said nodes to perform client services, said registration and launch service adapted to initiate client services on particular ones of said nodes in a cluster so that it appears to said client services that the entire cluster is substantially a single node containing said shared resource, wherein said registration and launch service chooses a node for each client service, registers the client service with that node, and notifies nodes presently in the cluster that the particular service is registered with the particular node.

36. In a multiprocessor system having multiple nodes, and a shared resource accessible to all nodes, a method of initiating client services on particular ones of said nodes in a cluster so that it appears to said client services that the entire cluster is substantially a single node containing said shared resource, said method including providing a registration and launch service which defines a computer process, said registration and launch service adapted to selecting one of said nodes for a client service, registering that client service with the selected node and launching that client service on the selected node according to an action parameter included with that client service in response to registering that client service with the selected node.

37. The method in claim 36 including transferring the client service to another node if that node leaves the cluster.

38. The method in claim 36 wherein said transferring includes relaunching the client service on the another node according to said action parameter.

39. In a multiprocessor system having multiple nodes, a shared resource accessible to all nodes and multiple interdependent levels of subsystems on each of said nodes, a method of combining particular ones of said nodes in a cluster so that it appears to users that the entire cluster is substantially a single node containing the shared resource, said method including taking said levels of said subsystems through node transitions in a manner that a particular level of said subsystems is taken through a node transition at all said ones of said nodes prior to taking a different level of said subsystems through the node transition.

40. The method of claim 39 wherein a higher level of said subsystems is dependent on the operation of a lower level of said subsystems.

41. The method of claim 40 including taking a lower level of said subsystems through a node transition prior to taking a higher level of said subsystems through the node transition if the node transition is a node joining the cluster.

42. The method of claim 40 including taking a lower level of said subsystems through a node transition prior to taking a higher level of said subsystems through the node transition if the node transition is a node ungracefully being forced from the cluster by other nodes.

43. The method of claim 40 including taking a higher level of said subsystems through a node transition prior to taking a lower level of said subsystems through the node transition if the node transition is a node gracefully leaving the cluster.

44. A computer usable medium in which program code is embodied, said program code defining an operating system for a multiprocessor cluster system having multiple nodes, and a shared resource accessible to all nodes, and including multiple interdependent levels of subsystems;

one of said subsystems being a cluster membership manager subsystem which defines a computer process for combining particular ones of said nodes in a cluster for that it appears to users that the entire cluster is substantially a single node containing the shared resource, said cluster membership manager subsystem adapted to taking all levels of subsystems through node transitions in a manner that a particular level of said subsystems is taken through a node transition at all said ones of said nodes prior to taking a different level of said subsystems through the node transition.