System and method for root cause analysis

• US 8,839,036 B2
• Filed: 12/30/2010
• Issued: 09/16/2014
• Est. Priority Date: 12/30/2010
• Status: Active Grant

• Current Assignee: Schneider Electric IT Corporation (Schneider Electric SE)
• Original Assignee: Schneider Electric IT Corporation (Schneider Electric SE)
• Inventors: Rymeski, Scott M., Ren, Tiegeng, Samson, Michael D.
• Primary Examiner(s): MASKULINSKI, MICHAEL C
• Application Number:
  US12/982,512
• Publication Number:
  US 20120173927A1
• Time in Patent Office: 1,356 Days
• Field of Search: 714/25, 714/26, 714/37
• US Class Current: 714/26
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0748 in a remote unit communicating with a single-box computer node experiencing an error/fault remote testing G06F11/2294
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36

Abstract

Systems and methods for determining the root cause of an event in a data center are presented. The system includes a data center management device coupled to a network and configured to receive an indication of the event from a physical infrastructure device via the network, determine a first generic cause model for the event by accessing an event cause model data store, determine a first event profile by adapting the first generic cause model to the data center using data center profile information stored in a data center profile data store and display a first probability that a potential cause defined by the first event profile is the root cause.

Claims

1. A system for determining a root cause of an event in a data center including a plurality of devices, the system comprising:

a data center management device coupled to a network, the data center management device including a memory and at least one processor coupled to the memory, the data center management device being configured to;

receive an indication of the event from a physical infrastructure device via the network;

determine a first generic cause model for the event by accessing an index that associates generic cause models with events, wherein the first generic cause model includes an indication of a relationship between the event and provision of a first data center resource;

determine a first event profile by adapting the first generic cause model to the data center using data center profile information stored in a data center profile data store, wherein the data center management device is configured to adapt the first generic cause model to the data center at least in part by identifying a first device coupled to the physical infrastructure device that is involved in provision of the first data center resource;

display a first probability that a potential cause defined by the first event profile is the root cause;

determine a second generic cause model for the event by accessing the index, the second generic cause model including an indication of a relationship between the event and provision of a second data center resource;

determine a second event profile by adapting the second generic cause model to the data center using data center profile information stored in the data center profile data store, wherein the data center management device is configured to adapt the second cause model to the data center at least in part by identifying a second device coupled to the physical infrastructure device that is involved in provision of the second data resource; and

display a second probability that another potential cause defined by the second event profile is the root cause.

2. The system according to claim 1, wherein the first data center resource is power.

3. The system according to claim 1, wherein the physical infrastructure device supplies the first data center resource to the first device.

4. The system according to claim 1, wherein the physical infrastructure device receives the first data center resource from the first device.

5. The system according to claim 1, wherein the first device is a component of the physical infrastructure device.

6. The system according to claim 1, wherein the data center management device is further configured to display at least one corrective action.

7. The system according to claim 6, wherein the data center management device is further configured to initiate the at least one correction action.

8. The system according to claim 1, wherein the first data center resource and the second data center resource are a same data center resource.

9. The system according to claim 1, wherein the second device is coupled to the physical infrastructure device via the first device.

10. The system according to claim 9, wherein the data center management device is further configured to:

display the first probability as a first link between two or more first nodes in a graph, each of the first nodes representing either the physical infrastructure device or another device of the plurality of devices; and

display the second probability as a second link between two or more second nodes in a graph, each of the second nodes representing either the physical infrastructure device or another device of the plurality of devices.

11. The system according to claim 10, wherein the data center management device is further configured to:

determine that the first probability is greater than the second probability; and

highlight the one or more first nodes responsive to determining that the first probability is greater than the second probability.

12. A method for determining a root cause of an event in a data center using a data center management device coupled to a network, the data center management device including memory and at least one processor coupled to the memory, the method comprising:

receiving indication of the event from a physical infrastructure device via the network;

determining a first generic cause model for the event by accessing an index that associates generic cause models with events, wherein the first generic cause model indicates a relationship between the event and provision of a first data center resource;

determining a first event profile by adapting the first generic cause model to the data center using data center profile information stored in a data center profile data store, wherein adapting the first generic cause model to the data center includes identifying a first device coupled to the physical infrastructure device that is involved in provision of the first data center resource;

displaying a first probability that a potential cause defined by the first event profile is the root cause;

determining a second generic cause model for the event by accessing the index, the second generic cause model indicating a relationship between the event and provision of a second data center resource;

determining a second event profile by adapting the second generic cause model to the data center using data center profile information stored in the data center profile data store, wherein adapting the second cause model to the data center includes identifying a second device coupled to the physical infrastructure device that is involved in provision of the second data resource; and

displaying a second probability that another potential cause defined by the second event profile is the root cause.

13. The method according to claim 12, wherein receiving the event includes receiving an event related to the provision of power.

14. The method according to claim 12, further comprising:

displaying the first probability as a first link between two or more first nodes in a graph, each of the first nodes representing either the physical infrastructure device or another device of a plurality of devices included in the data center; and

displaying the second probability as a second link between two or more second nodes in a graph, each of the second nodes representing either the physical infrastructure device or another device of the plurality of devices.

15. A non-transitory computer readable medium having stored thereon sequences of instruction for determining a root cause of an event in a data center including instructions that will cause at least one processor to:

receive an indication of the event from a physical infrastructure device via the network;

determine a first generic cause model for the event by accessing an index that associates generic cause models with events, wherein the first generic cause model includes an indication of a relationship between the event and provision of a first data center resource;

determine a first event profile by adapting the first generic cause model to the data center using data center profile information stored in a data center profile data store, wherein the instructions to determine the first event profile will further cause the at least one processor to adapt the first generic cause model to the data center at least in part by identifying a first device coupled to the physical infrastructure device that is involved in provision of the first data center resource;

display a first probability that a potential cause defined by the first event profile is the root cause

determine a second generic cause model for the event by accessing the index, the second generic cause model including an indication of a relationship between the event and provision of a second data center resource;

determine a second event profile by adapting the second generic cause model to the data center using data center profile information stored in the data center profile data store, wherein the instructions to determine the second event profile will further cause the at least one processor to adapt the second cause model to the data center at least in part by identifying a second device coupled to the physical infrastructure device that is involved in provision of the second data resource; and

display a second probability that another potential cause defined by the second event profile is the root cause.

16. The non-transitory computer readable medium according to claim 15, wherein the sequences of instruction include instructions that will further cause the at least one processor to display the first probability as first link between two or more first nodes in a graph, each of the first nodes representing either the physical infrastructure device or another device of a plurality of devices included in the data center.