SYSTEMS AND METHODS FOR SPATIALLY DISPLACED CORRELATION FOR DETECTING VALUE RANGES OF TRANSIENT CORRELATION IN MACHINE DATA OF ENTERPRISE SYSTEMS

• US 20160034328A1
• Filed: 07/29/2014
• Published: 02/04/2016
• Est. Priority Date: 07/29/2014
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Poola, Thyagaraju, Volchegursky, Vladimir, Karpuram, Venkata Ramana
• Granted Patent: US 9,658,910 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• 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 11/0754 by exceeding limits
• 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 11/34 Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
• CPC Class Codes: G06F 2201/81 Threshold

Abstract

Aspects of the present disclosure include systems and/or methods for detecting ranges of data that represent transient correlations in machine data corresponding to various hardware and/or software systems, such as enterprise systems employed by an information technology (“IT”) organization. In various aspects, the machine data may comprise one or more operational metrics that represent system performance, usage, and/or business activity of the enterprise system. The operational metrics may be used to identify operational issues within the enterprise system.

Claims

1. A method comprising:

generating, by one or more computing devices, a plurality of ranges of data based on a first dataset corresponding to a first metric describing a first performance of a system, wherein each range of data includes a first number of data points; and

for each range of data of the plurality of ranges of data;

identify a second number of data points from a second dataset corresponding to a second metric describing a second performance of the system, the second number of data points corresponding to the respective range of data; and

determine a correlation coefficient based on the first number of data points of the respective range of data and the second number of data points corresponding to the respective range of data;

determine a representative correlation coefficient based on the correlation coefficient; and

based on the representative correlation coefficient, storing an indication of whether or not the first dataset is correlated to the second dataset on the respective range of data of the representative correlation coefficient.

2. The method of claim 1, wherein each range of data includes a first edge defining a low boundary for the first number of data points and a second edge defining a high boundary for the first number of data points and wherein determining the representative correlation coefficient comprises:

increasing the first edge of the first number of data points to include a third number of data points within the respective range of data;

increasing the second number of data points from the second dataset to generate a fourth number of data points;

determining a second correlation coefficient based on the third number of data points and the fourth number of data points;

determining a first value, the first value being a sum of squares of correlation coefficients, based on the first correlation coefficient and the second correlation coefficient; and

when the first value is greater than the representative correlation coefficient, moving the first edge of the data points to include a fifth number of data points within the respective range of data.

3. The method of claim 2, further comprising:

when the first value is less than the representative correlation coefficient, identifying the first value as a new representative correlation coefficient.

4. The method of claim 1, wherein each range of data includes a first edge defining a low boundary for the number of data points and a second edge defining a high boundary for the number of data points and wherein determining the representative correlation coefficient comprises:

decreasing the second edge of the first number of data points to include a third number of data point within the respective range of data;

decreasing the second number of data points from the second dataset to generate a fourth number of data points;

determining a second correlation coefficient based on the third number of data points and the fourth number of data points;

determining a first value, the first value being a sum of squares of correlation coefficients, based on the first correlation coefficient and the second correlation coefficient; and

when the first value is greater than the representative correlation coefficient, moving the second edge of the data points to include a fifth number of data points within the respective range of data.

5. The method of claim 4, further comprising:

when the first value is less than the representative correlation coefficient, identifying the first value as a new representative correlation coefficient.

6. The method of claim 1, wherein each range of data of the plurality of ranges of data corresponds to a time interval during which the first number of data points were recorded and wherein the second number of data points corresponds to the respective range of data based on the time interval.

7. The method of claim 1, wherein the correlation coefficient is calculated using at least one of a Pearson or Spearman correlation algorithm and wherein the representative correlation coefficient is a maximum correlation coefficient of the correlation coefficients.

8. The method of claim 1, wherein the number of data points included within each range of data is randomly generated and includes at least enough data points to perform a statistical analysis.

9. The method of claim 1, further comprising:

based on the representative correlation coefficient of each discrete range of data, identifying a specific range of data from the plurality of ranges of data that identifies an acceptable correlation between the first metric and the second metric.

10. The method of claim 9, further comprising:

determining a p-value corresponding to the range of data; and

when the p-value proves a hypothesis of correlation between the first metric and the second metric, generating the indication.

11. The method of claim 1, further comprising:

based on the representative correlation coefficient of each discrete range of data, identifying specific ranges of data from the plurality of ranges of data that identifies an acceptable correlation between the first metric and the second metric; and

determining a plurality of p-values corresponding to the specific ranges of data; and

when at least one of the plurality of p-values proves a hypothesis of correlation between the first metric and the second metric, generating the indication.

12. A non-transitory computer readable medium encoded with instructions, the instructions executable by a processor, comprising:

generating a plurality of ranges of data based on a first dataset corresponding to a first metric describing a first performance of a system, wherein each range of data includes a first number of data points; and

for each range of data of the plurality of ranges of data;

identifying a second number of data points from a second dataset corresponding to a second metric describing a second performance of the system, the second number of data points corresponding to the respective range of data;

determining a correlation coefficient based on the first number of data points of the respective range of data and the second number of data points corresponding to the respective range of data; and

determining representative correlation coefficient based on the correlation coefficient; and

based on the representative correlation coefficient, storing an indication of whether or not the first dataset is correlated to the second dataset on the respective range of data of the representative correlation coefficient.

13. The non-transitory computer readable medium of claim 12, wherein each range of data includes a first edge defining a low boundary for the first number of data points and a second edge defining a high boundary for the first number of data points and wherein determining the representative correlation coefficient comprises:

increasing the first edge of the first number of data points to include a third number of data points within the respective range of data;

increasing the second number of data points from the second dataset to generate a fourth number of data points;

determining a second correlation coefficient based on the third number of data points and the fourth number of data points;

determining a first value, the first value being a sum of squares of correlation coefficients, based on the first correlation coefficient and the second correlation coefficient; and

when the first value is greater than the representative correlation coefficient, moving the first edge of the data points to include a fifth number of data points within the respective range of data.

14. The non-transitory computer readable medium of claim 12, further comprising:

when the first value is less than the representative correlation coefficient, identifying the first value as a new representative correlation coefficient.

15. The non-transitory computer readable medium of claim 12, wherein each range of data includes a first edge defining a low boundary for the number of data points and a second edge defining a high boundary for the number of data points and wherein determining the representative correlation coefficient comprises:

decreasing the second edge of the first number of data points to include a third number of data point within the respective range of data;

decreasing the second number of data points from the second dataset to generate a fourth number of data points;

determining a second correlation coefficient based on the third number of data points and the fourth number of data points;

determining a first value, the first value being a sum of squares of correlation coefficients, based on the first correlation coefficient and the second correlation coefficient; and

when the first value is greater than representative correlation coefficient, moving the second edge of the data points to include a fifth number of data points within the respective range of data.

16. The non-transitory computer readable medium of claim 15, further comprising:

when the first value is less than the representative correlation coefficient, identifying the first value as a new representative correlation coefficient.

17. The non-transitory computer readable medium of claim 12, wherein each range of data of the plurality of ranges of data corresponds to a time interval during which the first number of data points were recorded and wherein the second number of data points corresponds to the respective range of data based on the time interval.

18. The non-transitory computer readable medium of claim 12, wherein the correlation coefficient is calculated using at least one of a Pearson or Spearman correlation algorithm and wherein the representative correlation coefficient is a representative correlation coefficient of the correlation coefficients.

19. The non-transitory computer readable medium of claim 12, wherein the number of data points included within each range of data is randomly generated and includes at least enough data points to perform a statistical analysis.

20. The non-transitory computer readable medium of claim 12, further comprising:

based on the representative correlation coefficient of each discrete range of data, identifying a specific range of data from the plurality of ranges of data that identifies an acceptable correlation between the first metric and the second metric.

21. The non-transitory computer readable medium of claim 20, further comprising:

determining a p-value corresponding to the range of data; and

when the p-value proves a hypothesis of correlation between the first metric and the second metric, generating the indication.

22. The non-transitory computer readable medium of claim 12, further comprising:

based on the representative correlation coefficient of each discrete range of data, identifying specific ranges of data from the plurality of ranges of data that identifies an acceptable correlation between the first metric and the second metric; and

determining a plurality of p-values corresponding to the specific ranges of data; and

when at least one of the plurality of p-values proves a hypothesis of correlation between the first metric and the second metric, generating the indication.