Method and apparatus for analyzing and problem reporting in grid computing networks
View Patent ↗A method, device and computer product suitable for performing an analysis on a Grid Computer Network (GCN) is disclosed. The method comprises the steps of representing selected ones of a plurality of components and a relationship among the components in the GCN, providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event, and determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events. The step of representing the GCN, comprises the steps of creating at least one non-specific representation of the selected components, wherein the non-specific representations are selected from the group consisting of ComputeElement, GridService and GridServiceInstance, and creating at least one non-specification representation of relations along which the events propagate amongst the selected components, wherein the representations of relations are selected from the group consisting of ComposedOf, Part of, HostedBy, and HostsServices. Also disclosed is an apparatus comprising a processor and a computer program product providing computer instruction to a computing system for executing the method steps described herein.
1. A method for performing an analysis on a Grid Computer Network (GCN), the method comprising the steps of:
representing selected ones of a plurality of components and a relationship among the components in the GCN, wherein the Grid Computer Network is represented as plurality of logical domains;
providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event; providing, for each of the domains, a mapping between a plurality of observable events and a plurality of events for the components within the domain, wherein at least one of the observable events is associated with a component associated with at least two of the domains;
determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events; and
determining at least one likely event based on at least one of the plurality of observable events by determining a mismatch measure based on the values associated with the plurality of observable events and the plurality of events; wherein the Grid Computer Network is represented as a plurality of logical domains.
2. The method as recited in claim 1 , wherein the step of representing the GCN, comprises the steps of:
creating at least one non-specific representation of the selected components, wherein the non-specific representations are selected from the group consisting of: ComputeElement, GridService and GridServiceInstance; and
creating at least one non-specification representation of relations along which the events propagate amongst the selected components, wherein the representations of relations are selected from the group consisting of: ComposedOf, Part of, HostedBy, and HostsServices.
3. The method as recited in claim 1 , wherein at least one of the observable events is associated with at least one component associated with at least two of the domains.
4. A method for performing an analysis on a Grid Computer Network (GCN), the method comprising the steps of:
representing selected ones of a plurality of components and a relationship among the components in the GCN,
providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event;
determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events; wherein the analysis is selected from the group consisting of: fault detection, fault monitoring, performance, congestion, connectivity, interface failure, node failure, link failure, routing protocol error, routing control errors, and root-cause analysis.
5. An apparatus for performing an analysis on a Grid Computer Network (GCN), the apparatus comprising:
a processor in communication with a memory, the processor executing computer code for executing the steps of:
representing selected ones of a plurality of components and a relationship among the components;
creating at least one non-specific representation of the selected components, wherein the non-specific representations are selected from the group consisting of: ComputeElement, GridService and GridServiceInstance; and
creating at least one non-specification representation of relations along which the events propagate amongst the selected components, wherein the representations of relations are selected from the group consisting of: ComposedOf, Part of, HostedBy, and HostsServices;
providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event, and
determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events.
6. The apparatus as recited in claim 5 , wherein the analysis is selected from the group from consisting of: fault detection, fault monitoring, performance, congestion, connectivity, interface failure, node failure, link failure, routing protocol error, routing control errors, and root-cause analysis.
7. An apparatus for performing an analysis on a Grid Computer Network (GCN), the apparatus comprising:
a processor in communication with a memory, the processor executing computer code for executing the steps of:
representing selected ones of a plurality of components and a relationship among the components, wherein the Grid Computer Network is represented as plurality of logical domains;
providing, for each of the domains, a mapping between a plurality of observable events and a plurality of events for the components within the domain, wherein at least one of the observable events is associated with a component associated with at least two of the domains;
providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event, and
determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events.
8. The apparatus as recited in claim 7 , further comprising the step of:
determining at least one likely event based on at least one of the plurality of observable events by determining a mismatch measure based on the values associated with the plurality of observable events and the plurality of events.
9. The apparatus as recited in claim 8 , wherein at least one of the observable events is associated with at least one component associated with at least two of the domains.
10. A computer program product for performing an analysis on a Grid Computer Network (GCN), providing computer instruction to a computing system to execute the steps of:
representing selected ones of a plurality of components and a relationship among the components in the GCN, wherein the Grid Computer Network is represented as a plurality of logical domains;
providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event, wherein the mapping comprises providing, for each of the domains, a mapping between a plurality of observable events and a plurality of events for the components within the domain, wherein at least one of the observable events is associated with a component associated with at least two of the domains; and
determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events.
11. A computer program product for performing an analysis on a Grid Computer Network (GCN), providing computer instruction to a computing system to execute the steps of:
representing selected ones of a plurality of components and a relationship among the components in the GCN, wherein the step of representing the GCN, comprises the steps of:
creating at least one non-specific representation of the selected components, wherein the non-specific representations are selected from the group consisting of: ComputeElement, GridService and GridServiceInstance; and
creating at least one non-specification representation of relations along which the events propagate amongst the selected components, wherein the representations of relations are selected from the group consisting of: ComposedOf, Part of, HostedBy, and HostsServices
providing a mapping between a plurality of events and a plurality of observable events occurring among the components, wherein the mapping is represented as a value associating each event with each observable event, and
determining at least one likely causing event based on at least one of the plurality of observable events by determining a measure between each of a plurality of values associated with the plurality of observable events and the plurality of causing events.
12. The computer program product as recited in claim 10 , further comprising the step of:
determining at least one likely event based on at least one of the plurality of observable events by determining a mismatch measure based on the values associated with the plurality of observable events and the plurality of events.
13. The computer program product as recited in claim 10 , wherein at least one of the observable events is associated with at least one component associated with at least two of the domains.
14. The computer program product as recited in claim 10 , wherein the analysis is selected from the group consisting of: fault detection, fault monitoring, performance, congestion, connectivity, interface failure, node failure, link failure, routing protocol error, routing control errors, and root-cause analysis.