INFORMATION PROCESSING SYSTEM MONITORING APPARATUS, MONITORING METHOD, AND MONITORING PROGRAM

• US 20150254125A1
• Filed: 10/29/2013
• Published: 09/10/2015
• Est. Priority Date: 01/11/2013
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Kakui, Kentaro
• Granted Patent: US 9,588,833 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• 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 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/323 Visualisation of programs or trace data
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Provided is a method of detecting a warning sign indicating the occurrence of a system failure targeting a plurality of varied information processing systems. A system for monitoring a plurality of monitored systems receives measured values regarding a plurality of indexes, designates, from a plurality of prediction models, a prediction model for predicting a future measured value from the measured value that was received regarding a reference index, predicts a predicted value of the reference index based on the designated prediction model, generates or updates a Bayesian network targeting the reference index and a target index, and calculates a probability that a measured value of the target index will become a predetermined value or fall within a predetermined value range.

Claims

1. A service system which is configured with one or more computers and performs a failure prediction of a plurality of monitored systems,

wherein the one or more computers:

(1) receive measured values regarding a plurality of indexes from the monitored systems;

(2) designate a prediction model for predicting a future measured value from the received measured values, from a plurality of prediction models, regarding a reference index which is a part of the plurality of indexes;

(3) predict a predicted value of the reference index based on the designated prediction model;

(4) generate or update a Bayesian network targeting the reference index and a target index which is an index that is a part of the plurality of indexes but different from the reference index; and

(5) calculate a probability that a measured value of the target index will become a predetermined value or fall within a predetermined value range based on the predicted value of the reference index, and the Bayesian network.

2. The service system according to claim 1,

wherein the one or more computers generate the Bayesian network based on the received measured value of the reference index, and the measured value of the target index.

3. The service system according to claim 1,

wherein the measured value is data which is a quantification of an execution status of an application program executed by the monitored system.

4. The service system according to claim 1,

wherein the one or more computers:

(6) receive an input designating the reference index from the plurality of indexes for each of the plurality of monitored systems; and

(7) can set a different index for each monitored system as a reference index by managing, as a set, an identifier of the monitored system, the designated prediction model and the reference index.

5. The service system according to claim 1,

wherein the one or more computers:

(2a) receive an input which designates the prediction model.

6. The service system according to claim 1,

wherein the one or more computers:

(4a) generate a plurality of Bayesian networks, and store the generated Bayesian networks in a memory or a storage of the one or more computers; and

(8) receive an input which designates, from the plurality of Bayesian networks, the Bayesian network to be used in the probability calculation of (5).

7. The service system according to claim 1,

wherein the one or more computers:

(9) receive an input which designates the predetermined value or value range to be used in the probability calculation.

8. The service system according to claim 1,

wherein the one or more computers control an execution priority of the processing of (4) and (5).

9. The service system according to claim 8,

wherein the execution priority is effective even between different monitored systems.

10. The service system according to claim 8,

wherein the execution priority is controlled based on at least one among:

a target time of the probability calculation;

number of indexes of the monitored system; and

level of change in past measured values.

11. A program to be executed by one or more computers configuring a service system which performs a failure prediction of a plurality of monitored systems,

wherein the program:

(1) receives measured values regarding a plurality of indexes from the monitored systems;

(2) designates a prediction model for predicting a future measured value from the received measured values, from a plurality of prediction models, regarding a reference index which is a part of the plurality of indexes;

(3) predicts a predicted value of the reference index based on the designated prediction model;

(4) generates or updates a Bayesian network targeting the reference index, and a target index which is an index that is a part of the plurality of indexes but different from the reference index; and

(5) calculate a probability that a measured value of the target index will become a predetermined value or fall within a predetermined value range based on the predicted value of the reference index, and the Bayesian network.

12. The program according to claim 11,

wherein the program generates the Bayesian network based on the received measured value of the reference index, and the measured value of the target index.

13. The program according to claim 11,

wherein the measured value is data which is a quantification of an execution status of an application program executed by the monitored system.

14. The program according to claim 11,

wherein the program:

(6) receives an input designating the reference index from the plurality of indexes for each of the plurality of monitored systems; and

(7) can set a different index for each monitored system as a reference index by managing, as a set, an identifier of the monitored system, the designated prediction model, and the reference index.

15. The program according to claim 11,

wherein the program:

(2a) receives an input which designates the prediction model.

16. The program according to claim 11,

wherein the program:

(4a) generates a plurality of Bayesian networks, and stores the generated Bayesian networks in a memory or a storage of the one or more computers; and

(8) receives an input which designates, from the plurality of Bayesian networks, the Bayesian network to be used in the probability calculation of (5).

17. The program according to claim 11,

wherein the program:

(9) receives an input which designates the predetermined value or value range to be used in the probability calculation.

18. The program according to claim 11,

wherein the program controls an execution priority of the processing of (4) and (5).

19. The program according to claim 18,

wherein the execution priority is effective even between different monitored systems.

20. The program according to claim 18,

wherein the execution priority is controlled based on at least one among:

a target time of the probability calculation;

number of indexes of the monitored system; and

level of change in past measured values.