SYSTEM AND METHOD TO PREDICTIVELY SERVICE AND SUPPORT THE SOLUTION

• US 20190324841A1
• Filed: 04/24/2018
• Published: 10/24/2019
• Est. Priority Date: 04/24/2018
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Inventors: Patel, Dharmesh M., Ali, Rizwan, Chaganti, Ravikanth
• Granted Patent: US 10,795,756 B2
• Time in Patent Office: Days
• 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 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition

Abstract

A system state monitor for managing a distributed system includes a persistent storage and a processor. The persistent storage includes a heuristically derived knowledge base. The processor performs deployment-level monitoring of deployments of the distributed system and identifies a common component failure of components of the deployments based on the deployment-level monitoring. In response to identifying the common component failure, the processor identifies impacted computing devices each hosting a respective component of the components; obtains deployment level state information from each of the impacted computing devices; identifies an iterative set of outcome driven corrective actions based on the obtained deployment level state information and the heuristically derived knowledge base; and initiates a computing device correction on an impacted computing device of the impacted computing devices using the iterative set of outcome driven corrective actions to obtain a corrected computing device.

Claims

1. A system state monitor for managing a distributed system, comprising:

a persistent storage comprising a heuristically derived knowledge base; and

a processor programmed to;

perform deployment-level monitoring of deployments of the distributed system;

identify a common component failure of a plurality of components of the deployments based on the deployment-level monitoring;

in response to identifying the common component failure;

identify a plurality of impacted computing devices each hosting a respective component of the plurality of components;

obtain deployment level state information from each of the plurality of impacted computing devices;

identify an iterative set of outcome driven corrective actions based on the obtained deployment level state information and the heuristically derived knowledge base; and

initiate a computing device correction on an impacted computing device of the plurality of impacted computing devices using the iterative set of outcome driven corrective actions to obtain a corrected computing device.

2. The support engine of claim 1, wherein the iterative set of outcome driven corrective actions comprises:

a primary corrective action associated with the system level state information.

3. The support engine of claim 2, wherein the iterative set of outcome driven corrective actions further comprises:

a first plurality of primary outcomes associated with the primary corrective action,

wherein each primary outcome of the first plurality of primary outcomes specifies different system level state information.

4. The support engine of claim 3, wherein the iterative set of outcome driven corrective actions further comprises:

a secondary corrective action associated with a primary outcome of the plurality of primary outcomes.

5. The support engine of claim 4, wherein the heuristically derived knowledge base comprises:

the primary corrective action;

the plurality of primary outcomes; and

the secondary corrective action.

6. The support engine of claim 1, wherein a first component of the plurality of components is hosted by a computing device of a first deployment of the deployments and a second component of the plurality of components is hosted by a computing device of a second deployment of the deployments.

7. The support engine of claim 1, wherein a first deployment of the deployments provides data storage services to the distributed system.

8. The support engine of claim 7, wherein a second deployment of the deployments provides computation services to the distributed system.

9. The support engine of claim 7, wherein a second deployment of the deployments provides caching services to the distributed system.

10. The support engine of claim 1, wherein the processor is further programmed to:

make a determination that the corrected computing device is in a malfunctioning state after initiating the iterative set of outcome driven corrective actions; and

in response to the determination, update the heuristically derived knowledge base.

11. The support engine of claim 10, wherein updating the heuristically derived knowledge base comprises:

adding a new outcome based on the malfunctioning state; and

adding a new corrective action associated with the new outcome.

12. The support engine of claim 1, wherein each impacted computing device is a member of the deployments.

13. A method for managing a distributed system, comprising:

performing deployment-level monitoring of deployments of the distributed system;

identifying a common component failure of a plurality of components of the deployments based on the deployment-level monitoring;

in response to identifying the common component failure;

identifying a plurality of impacted computing devices each hosting a respective component of the plurality of components;

obtaining deployment level state information from each of the plurality of impacted computing devices;

identifying an iterative set of outcome driven corrective actions based on the obtained deployment level state information and a heuristically derived knowledge base; and

initiating a computing device correction on an impacted computing device of the plurality of impacted computing devices using the iterative set of outcome driven corrective actions to obtain a corrected computing device.

14. The method of claim 13, wherein the iterative set of outcome driven corrective actions comprises:

a primary corrective action associated with the system level state information.

15. The method of claim 14, wherein the iterative set of outcome driven corrective actions further comprises:

a first plurality of primary outcomes associated with the primary corrective action,

wherein each primary outcome of the first plurality of primary outcomes specifies different system level state information.

16. The method of claim 15, wherein the iterative set of outcome driven corrective actions further comprises:

a secondary corrective action associated with a primary outcome of the plurality of primary outcomes.

17. The method of claim 16, wherein the heuristically derived knowledge base comprises:

the primary corrective action;

the plurality of primary outcomes; and

the secondary corrective action.

18. The method of claim 13, wherein a first component of the plurality of components is hosted by a computing device of a first deployment of the deployments and a second component of the plurality of components is hosted by a computing device of a second deployment of the deployments.

19. A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for managing a distributed system, the method comprising:

performing deployment-level monitoring of deployments of the distributed system;

identifying a common component failure of a plurality of components of the deployments based on the deployment-level monitoring;

in response to identifying the common component failure;

identifying a plurality of impacted computing devices each hosting a respective component of the plurality of components;

obtaining deployment level state information from each of the plurality of impacted computing devices;

identifying an iterative set of outcome driven corrective actions based on the obtained deployment level state information and a heuristically derived knowledge base; and

initiating a computing device correction on an impacted computing device of the plurality of impacted computing devices using the iterative set of outcome driven corrective actions to obtain a corrected computing device.

20. The non-transitory computer readable medium of claim 19, wherein a first component of the plurality of components is hosted by a computing device of a first deployment of the deployments and a second component of the plurality of components is hosted by a computing device of a second deployment of the deployments.