Software system and methods for analyzing the performance of a server

• US 6,898,556 B2
• Filed: 09/26/2003
• Issued: 05/24/2005
• Est. Priority Date: 08/06/2001
• Status: Expired due to Term

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Mercury Interactive Corp. (HP Inc.)
• Inventors: Landan, Gideon, Smocha, Itzhak
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Barbee, Manuel L
• Application Number:
  US10/672,445
• Publication Number:
  US 20040059544A1
• Time in Patent Office: 606 Days
• Field of Search: 702/108, 702/119, 702/122, 702/123, 702/176, 702/179, 702180-183, 702/186, 702/187, 702/188, 709/224, 714/43, 714/47
• US Class Current: 702/186
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3457 Performance evaluation by simulation
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet
• CPC Class Codes: G06F 2201/885 Monitoring specific for caches

Abstract

A system for monitoring and evaluating the performance of a server comprises a component that monitors a plurality of performance metrics of the server over a period of time while the server responds to requests from clients. An analysis component analyzes the collected performance metric data values to evaluate degrees to which specific performance metrics are related. Causal relationships and dependencies that affect the performance of the server are thereby identified.

Claims

1. A computer-implemented method of analyzing the performance of a server, the method comprising:

monitoring a first performance metric of the server over a period of time to generate a series of values of the first performance metric, wherein the server responds to requests from clients during said period of time;

monitoring a second performance metric of the server over said period of time to generate a series of values of the second performance metric;

programmatically identifying a time segment in which the values of the first and second performance metrics are significant, and identifying the time segment to a user, wherein the time segment is shorter than said period of time; and

performing an automated correlation analysis that is substantially limited to the time segment to measure a degree of correlation between the first and second performance metrics within the time segment.

2. The method of claim 1, wherein the automated correlation analysis is performed in response to user selection of the time segment.

3. The method of claim 1, wherein identifying the time segment to the user comprises displaying a graph that corresponds to the time segment and depicts values of the first and second performance metrics therein.

4. The method of claim 1, further comprising providing an option for the user to select the time segment and to initiate the automated correlation analysis.

5. The method of claim 1, wherein programmatically identifying a time segment comprises identifying a transition point at which behavior of a performance metric changes, and using the transition point to define a boundary of said time segment.

6. The method of claim 1, further comprising compensating for a time offset that exists between the series of values of the first performance metric and the series of values of the second performance metric.

7. The method of claim 6, further comprising detecting said time offset by identifying corresponding but offset transition points in the series of values of the first performance metric and the series of values of the second performance metric.

8. The method of claim 1, further comprising applying a re-sampling method to the series of values of the first performance metric to facilitate a comparison with the series of values of the second performance metric.

9. The method of claim 1, wherein the first performance metric is a server response time.

10. The method of claim 9, wherein the second performance metric represents a current load on the server.

11. The method of claim 9, wherein the second performance metric represents server transactions executed per unit time.

12. The method of claim 1, further comprising applying a controlled load to the server during said period of time.

13. A computer system programmed to perform the method of claim 1.

14. A computer program that embodies the method of claim 1 represented in a computer-readable medium.

15. A computer-implemented method for analyzing the performance of a server, the method comprising:

monitoring a plurality of performance metrics of the server over a period of time to generate a collection of performance metric data values;

performing a sampling analysis of the collection of performance metric data values to identify a significant portion thereof;

analyzing performance metric data values within the significant portion to measure, for at least one pair of said performance metrics, a degree to which the pair of performance metrics are correlated; and

identifying to a user a segment in which the performance metric data values are significant, and providing to the user an option to perform an automated correlation analysis on the segment.

16. The method of claim 15, wherein monitoring the plurality of performance metrics comprises storing each performance metric data value in association with a time index.

17. The method of claim 15, wherein the plurality of performance metrics include performance metrics of at least the following types:

(a) server transactions executed per unit time, (b) transaction response time.

18. The method of claim 15, wherein the pair of performance metrics includes a first performance metric that represents a response time of the server and a second performance metric that represents a load on the server.

19. The method of claim 15, wherein performing the sampling analysis comprises separately analyzing a series of data values of each performance metric.

20. The method of claim 15, wherein performing the sampling analysis comprises evaluating whether collected performance metric data values exhibit a statistically significant trend.

21. The method of claim 15, wherein performing the sampling analysis comprises determining whether a performance metric is uninformative.

22. The method of claim 15, wherein performing the sampling analysis comprises determining whether a sufficient number of data values have been collected for a given performance metric to perform a statistically significant analysis.

23. The method of claim 15, wherein performing the sampling analysis comprises identifying at least one segment of time for which the performance metric data values are sufficient for performing a statistically meaningful correlation analysis.

24. The method of claim 15, wherein analyzing the performance metric data values comprises calculating a correlation coefficient that represents a degree to which the pair of performance metrics are correlated.

25. The method of claim 24, wherein the correlation coefficient is a Pearson correlation coefficient.

26. The method of claim 24, wherein the correlation coefficient is a coefficient of determination.

27. The method of claim 24, further comprising determining whether a meaningful relationship exists between the pair of performance metrics at least in part by comparing the correlation coefficient to upper threshold and a lower threshold.

28. The method of claim 15, further comprising detecting and compensating for a time offset that exists between the pair of performance metrics prior to measuring the degree to which the pair of performance metrics are correlated.

29. The method of claim 15, further comprising applying a re-sampling method to a series of data values of a performance metric of said pair.

30. The method of claim 15, further comprising applying a controlled load to the server during said period of time.

31. A computer system programmed to perform the method of claim 15.

32. A computer program that embodies the method of claim 15 represented in a computer-readable medium.

33. A computer-implemented method for analyzing the performance of a server, the method comprising:

monitoring a plurality of performance metrics of the server over a period of time to generate a collection of performance metric data values;

performing a sampling analysis of the collection of performance metric data values to identify a significant portion thereof;

analyzing performance metric data values within the significant portion to measure, for at least one pair of said performance metrics, a degree to which the pair of performance metrics are correlated; and

dividing the collection of performance metric data values into multiple significant segments based on said sampling analysis, each segment representing a different window of time.

34. A system for facilitating the analysis of the operation of a server, the system comprising:

a data collection component that collects data values of each of a plurality of performance metrics reflective of the performance of the server;

an analysis component that analyzes the data values at least in-part by generating correlation coefficients for specific pairs of the performance metrics to identify performance metrics that are related; and

a display component that identifies to a user segments of performance metric data values that are deemed significant, and provides an option for the user to select a segment on which to perform an automated correlation analysis.

35. The system of claim 34, wherein the analysis component evaluates whether a meaningful relationship exists between a pair of performance metrics at least in-part by comparing a correlation coefficient associated with the pair to a lower threshold value and an upper threshold value.

36. The system of claim 34, wherein the analysis component calculates Pearson correlation coefficients for the pairs of performance metrics.

37. The system of claim 34, wherein the analysis component calculates correlation coefficients of determination for the pairs of performance metrics.

38. The system of claim 34, wherein the analysis component uses a clustering algorithm to group together performance metrics that are deemed related.

39. The system of claim 34, wherein the analysis component generates a tree indicating performance metrics that, based on a clustering analysis, are deemed to demonstrate similar behavior.

40. The system of claim 34, wherein the analysis component detects and compensates for offsets in time between transition points in a pair of performance metrics.

41. The system of claim 34, wherein the analysis component performs a sampling analysis on a series of the data values to identify significant portions within said series.

42. The system of claim 34, wherein the data collection component associates a time value with each data value of each performance metric.

43. The system of claim 34, wherein the plurality of performance metrics include performance metrics indicative of at least:

(a) server transactions executed per unit time, and (b) server response time.

44. The system of claim 34, further comprising a load generation component that applies a controlled load to the server during collection of performance metric data values by the data collection component.

45. A system for facilitating the analysis of the operation of a server, the system comprising:

a data collection component that collects data values of each of a plurality of performance metrics reflective of the performance of the server over a period of time to generate a collection of performance metric data values; and

an analysis component that analyzes the collection of performance metric data values at least in-part by a performing a sampling analysis to identify significant portions thereof, and by analyzing the significant portions to measure degrees to which specific pairs of performance metrics are correlated;

wherein the analysis component identifies transition points at which behaviors of the performance metrics change, and uses the transition points as boundaries for dividing the collection of performance metric data values into multiple segments.

46. The system of claim 45, wherein the data collection component associates each performance metric data value with a time index.

47. The system of claim 45, wherein the analysis component performs the sampling analysis at least in-part by separately analyzing a series of data values of each performance metric.

48. The system of claim 45, wherein the analysis component performs the sampling analysis at least in-part by evaluating whether collected performance metric data values exhibit a statistically significant trend.

49. The system of claim 45, wherein the analysis component performs the sampling analysis at least in-part by determining whether a performance metric is uninformative.

50. The system of claim 45, wherein the analysis component performs the sampling analysis at least in-part by determining whether a sufficient number of data values have been collected for a given performance metric to perform a statistically significant analysis.

51. The system of claim 45, wherein the analysis component performs the sampling analysis at least in-part by identifying at least one segment of time for which the performance metric data values are sufficient for performing a statistically meaningful correlation analysis.

52. The system of claim 45, wherein the analysis component further re-samples a series of data values of a first performance metric to facilitate a comparison with a series of data values of a second performance metric.

53. The system of claim 45, wherein the analysis component further detects and compensates for a time offset between a pair of performance metrics.

54. The system of claim 45, wherein the analysis component analyzes the significant portions at least in-part by generating correlation coefficients for specific pairs of the performance metrics.

55. The system of claim 54, wherein the correlation coefficients include Pearson correlation coefficients.

56. The system of claim 54, wherein the analysis component evaluates whether a meaningful relationship exists between a pair of performance metrics at least in-part by comparing a correlation coefficient associated with the pair to a lower threshold value and an upper threshold value.

57. The system of claim 45, wherein the plurality of performance metrics include performance metrics indicative of at least:

(a) server transactions executed per unit time, and (b) transaction response time.

58. The system of claim 45, further comprising a load generation component that applies a controlled load to the server during collection of performance metric data values by the data collection component.

59. The system of claim 45, further comprising a display component that identifies said significant portions to a user.

60. The system of claim 59, wherein the display component provides a user option to select a significant portion on the analysis component is to perform a correlation analysis.