Fault tolerance framework for networks of nodes

• US 8,527,622 B2
• Filed: 10/12/2007
• Issued: 09/03/2013
• Est. Priority Date: 10/12/2007
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Moreira Sa de Souza, Luciana
• Primary Examiner(s): DAILEY, THOMAS J
• Application Number:
  US11/871,616
• Publication Number:
  US 20090097397A1
• Time in Patent Office: 2,153 Days
• Field of Search: 709/223, 709/224, 709/230, 714/13
• US Class Current: 709/224
• CPC Class Codes: H04L 41/0233 Object-oriented techniques, for representation of network management data, e.g. common object request broker architecture [CORBA]
• CPC Class Codes: H04L 41/0273 using web services for network management, e.g. simple object access protocol [SOAP]
• CPC Class Codes: H04L 41/0636 based on a decision tree analysis
• CPC Class Codes: H04L 41/0663 Performing the actions predefined by failover planning, e.g. switching to standby network elements
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
• CPC Class Codes: H04L 69/08 Protocols for interworking; Protocol conversion
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

In some implementations, a first message handler may be configured to receive first network-related data associated with a first network of nodes, the first network of nodes using a first communications protocol. A second message handler may be configured to receive second network-related data associated with a second network of nodes, the second network of nodes using a second communications protocol. A message transport system may be configured to receive the first network-related data and the second network-related data and further configured to route the first network-related data and the second network-related data in a common protocol, and a fault manager may be configured to receive the network-related data in the common protocol and configured to determine a fault associated with an operation of one or more of the first network of nodes and the second network of nodes, based on the network-related data in the common protocol.

Claims

1. A system including instructions recorded on a non-transitory computer-readable storage medium and executable by at least one processor, the system comprising:

the at least one processor;

a first message handler configured to cause the at least one processor to receive first network-related data using a first communications protocol, the first network-related data being associated with a first network of nodes, the nodes of the first network of nodes communicating with one another within the first network using the first communications protocol;

a second message handler configured to cause the at least one processor to receive second network-related data using a second communications protocol, the second network-related data being associated with a second network of nodes, the nodes of the second network of nodes communicating with one another within the second network using the second communications protocol;

a message transport system configured to cause the at least one processor to receive the first network-related data and the second network-related data and further configured to route the first network-related data and the second network-related data in a common communications protocol; and

a fault manager configured to;

construct and maintain a state model that stores events occurring in the first network of nodes and the second network of nodes, stores related state information, and stores network metadata for the first network of nodes and the second network of nodes,

cause the at least one processor to receive the network-related data in the common communications protocol, and

determine a fault associated with an operation of one or more of the first network of nodes and the second network of nodes, based on the network-related data in the common protocol and on the state model,

cause the at least one processor to determine a recovery method for recovering an operation of the networks of nodes despite the fault, the recovery method being determined from among a plurality of recovery methods including one or more of triggering a workflow, performing a sensor value fusion, and deploying a service within a node of the networks of nodes; and

cause the at least one processor to implement the determined recovery method; and

a code distribution manager configured to determine a target node not associated with the fault within the first network of nodes and to determine a service executable for deploying a service in response to the fault and based on a mapping of the service to the target node as part of implementing the determined recovery method when the recovery method includes deploying a service; and

a service injector configured to deploy the service executable to the target node for continued execution thereon to thereby recover the operation, wherein the service injector is selected from a plurality of service injectors as being compatible with the first communications protocol.

2. The system of claim 1 wherein the first message handler comprises:

a native message handler in communication with a message bridge to receive the first network-related data, the message bridge being in communication with at least one device of the first network of nodes using the first communications protocol,

wherein the native message handler is configured to translate the first network-related data from the first communications protocol to the common communications protocol.

3. The system of claim 1 wherein the second message handler comprises:

a gateway message handler in communication with a service gateway to receive the second network-related data, the service gateway being in communication with at least one device of the second network of nodes and being compatible with the second communications protocol,

wherein the gateway message handler is configured to receive the second network-related data in the common communications protocol from the service gateway for forwarding to the message transport system.

4. The system of claim 1 wherein the message transport system comprises a notification broker configured to publish at least some of the first network-related data and/or the second network-related data using the common communications protocol, for determination of the fault based thereon.

5. The system of claim 1 wherein the message transport system comprises:

a request processor that is configured to;

receive a request for additional network-related data,

select the first network as being configured to provide the additional network-related data, based on a content of the request, and

forward the request to the first message handler.

6. The system of claim 1 wherein the events stored in the state model include pre-fault events.

7. The system of claim 1 wherein the fault manager includes a fault diagnosis manager configured to determine that an event associated with at least the first network of nodes is associated with the fault.

8. The system of claim 7 wherein the fault diagnosis manager comprises:

a fault detector configured to determine a type of fault and a fault detection algorithm associated with the type of fault, and to detect the fault based on the fault detection algorithm and the event.

9. The system of claim 8 wherein the fault diagnosis manager comprises:

a fault isolator configured to receive the detected fault from the fault detector and configured to determine a cause of the fault relative to the networks of nodes.

16. The system of claim 1, wherein the nodes of the first network of nodes include first sensors, and wherein the nodes of the first network of nodes are configured to execute a first collaborative sensing function using the first communications protocol, and wherein the nodes of the second network of nodes include second sensors, and wherein the nodes of the second network of nodes are configured to execute a second collaborative sensing function using the second communications protocol.

10. A system including instructions recorded on a non-transitory computer-readable storage medium and executable by at least one processor, the system comprising:

the at least one processor;

a middleware layer configured to cause the at least one processor to communicate with a plurality of networks of nodes and configured to cause the at least one processor to communicate with at least one back-end application, the middleware layer including

a platform abstraction layer configured to;

cause the at least one processor to receive, from the plurality of networks of nodes, first network-related data using a first communications protocol from a first network of nodes and second network-related data using a second communications protocol from a second network of nodes, wherein the first communications protocol is used by the nodes of the first network for communicating with one another within the first network and the second communications protocol is used by the nodes of the second network for communicating with one another within the second network, and

provide the first network-related data and the second network-related data in a common protocol; and

a fault management layer configured to;

cause the at least one processor to construct and maintain a state model that stores events occurring in the first network of nodes and the second network of nodes, stores related state information, and stores network metadata for the first network of nodes and the second network of nodes,

cause the at least one processor to receive the first network-related data and the second network-related data in the common protocol, and

cause the at least one processor to determine a fault associated with an operation of the plurality of networks, based on the network-related data in the common protocol and on the state model,

cause the at least one processor to determine a recovery method for responding to the fault, and

cause the at least one processor to notify the platform abstraction layer of the recovery method, wherein the platform abstraction layer is further configured to;

cause the at least one processor to determine a target node not associated with the fault within the first network of nodes when the recovery method includes deploying a service,

cause the at least one processor to select a service injector from a plurality of service injectors, the selected service injector being compatible with the first communications protocol, and

cause the at least one processor to deploy the service, using the selected service injector, to the target node for execution thereon to thereby recover the fault.

11. The system of claim 10 wherein the platform abstraction layer comprises a plurality of message handlers, each message handler being configured to translate a corresponding one of at least the first communications protocol and the second communications protocol into the common protocol.

12. The system of claim 10 wherein the fault management layer comprises:

a fault diagnosis manager configured to determine that an event associated with the operation is associated with the fault, and configured to determine a cause of the fault based in part on the event.

20. The system of claim 10 wherein as part of determining the fault, the fault management layer is further configured to cause the at least one processor to:

determine a type of failure; and

request additional information from the state model based on the type of failure.

13. A method comprising:

receiving network-related data associated with a plurality of networks of nodes at one of a plurality of message handlers, the plurality of message handlers each associated with a corresponding network of nodes and a corresponding communications protocol that is used by the corresponding network of nodes to conduct in-network communications within and among the corresponding nodes thereof;

translating the network-related data from the corresponding communications protocol into a common communications protocol;

providing the network-related data in the common communications protocol to a state model that stores state information related to the plurality of networks of nodes and network metadata for the plurality of networks of nodes;

diagnosing a fault associated with an operation of the plurality of networks of nodes, based on the state model, the operation being associated with a source node of a first network of nodes within the plurality of networks; and

recovering the fault by;

determining a recovery method for recovering the operation despite the fault,

determining a target node not associated with the fault within the first network of nodes when the recovery method includes deploying a service,

selecting a service injector from a plurality of service injectors, the selected service injector being compatible with a communications protocol corresponding to the source node, and

deploying, using the selected service injector, the service to the target node for execution on the target node to recover the fault.

14. The method of claim 13 wherein providing the network-related data in the common communications protocol comprises:

publishing a notification of the network-related data to the state model, based on a content of the network-related data and on a subscription of the state model to the content.

15. The method of claim 13 wherein diagnosing the fault comprises:

determining an event associated with the networks of nodes from the state model;

determining a type of failure; and

determining that the event is associated with the type of failure.

17. The method of claim 15, wherein diagnosing the fault further includes:

requesting additional information from the state model based on the type of failure.

18. The method of claim 17, wherein diagnosing the fault further includes:

determining a fault recovery procedure based on the type of failure, a location of failure, and a time of failure; and

identifying the target node as able to assume functionality for the source node.

19. The method of claim 13, wherein recovering the fault further includes:

identifying the service as a suitable replacement for a service that has been diagnosed as contributing to the fault.

21. The method of claim 13, wherein diagnosing the fault comprises:

determining that an event associated with the operation is associated with the fault, and

determining a cause of the fault based in part on the event.