Performance Metrics of a Computer System

• US 20150046757A1
• Filed: 08/11/2014
• Published: 02/12/2015
• Est. Priority Date: 08/12/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Brew, Anthony T., Flores Snchez, Laura, Murphy, Paul J., Palmia, Michele, Ward, Anthony W.
• Granted Patent: US 9,524,223 B2
• Time in Patent Office: Days
• US Class Current: 714/47.3
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis

Abstract

Identifying an inter-relationship between performance metrics of a computer system. It is proposed to convert performance metric signals, which represent variations of performance metrics over time, into quantized signals having a set of allowable discrete values. The quantized signals are compared to detect a correlation based on the timing of variations quantized signals. An inter-relationship between the performance metrics may then be identified based on a detected correlation between the quantized signals.

Claims

1. A method of identifying an inter-relationship between performance metrics of a computer system, the method comprising steps of:

obtaining a first metric signal for a first performance metric, the first metric signal representing a variation of the first performance metric over time;

obtaining a second metric signal for a second performance metric, the second metric signal representing a variation of the second performance metric over time;

converting the first metric signal into a first quantized signal having a first set of allowable discrete values;

converting the second metric signal into a second quantized signal having a second set of allowable discrete values;

detecting a correlation between the first and second quantized signals based on a timing of variations in the first and second quantized signals; and

identifying an inter-relationship between the first and second performance metrics based on the detected correlation between the first and second quantized signals.

2. The method of claim 1, wherein the first and second metric signals represent a variation of the first and second performance metrics, respectively, over a common time period.

3. The method of claim 1, wherein the first set of allowable discrete values comprises first and second values and wherein converting the first metric signal into a first quantized signal comprises approximating a value of the first metric signal at a time instant to the second value if it exceeds a predetermined threshold.

4. The method of claim 1, wherein at least one of the first and second performance metrics comprise a user-definable key performance indicator (KPI).

5. The method of claim 1, further comprising a step of:

generating a graphical representation of the identified inter-relationship between the first and second performance metrics.

8. A computer program product for identifying an inter-relationship between performance metrics of a computer system, wherein the computer program product comprises a non-transitory computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code configured to perform all of the steps of claim 1.

9. A computer system comprising:

a computer program product according to claim 8; and

one or more processors adapted to perform all of the steps of claim 1.

6. A method of monitoring a computer system comprising steps of:

identifying an inter-relationship between first and second performance metrics of the computer system;

determining a dependency between the first and second performance metrics based on the identified inter-relationship between the first and second performance metrics; and

based on the determined dependency, monitoring only one of the first and second performance metrics to detect an occurrence of an anomaly in the monitored performance metric.

7. The method of claim 6, further comprising steps of:

detecting the occurrence of an anomaly in the monitored performance metric; and

communicating a signal warning of potential abnormality occurrence in the other performance metric.

10. A system comprising:

a computer system;

a first detection unit adapted to obtain a first metric signal for a first performance metric of the computer system, the first metric signal representing a variation of the first performance metric over time;

a second detection unit adapted to obtain a second metric signal for a second performance metric of the computer system, the second metric signal representing a variation of the second performance metric over time;

a quantizing unit adapted to convert the first metric signal into a first quantized signal having a first set of allowable discrete values and to convert the second metric signal into a second quantized signal having a second set of allowable discrete values; and

a data analysis unit adapted to detect a correlation between the first and second quantized signals based on a timing of variations in the first and second quantized signals, and to identify an inter-relationship between the first and second performance metrics based on the detected correlation between the first and second quantized signals.

11. The system of claim 10 wherein the first and second metric signals represent a variation of the first and second performance metrics, respectively, over a common time period.

12. The system of claim 10, wherein the first set of allowable discrete values comprises first and second values and wherein the quantizing unit is adapted to convert the first metric signal into a first quantized signal by approximating a value of the first metric signal at a time instant to the second value if it exceeds a predetermined threshold.

13. The system of claim 10, wherein at least one of the first and second performance metrics comprise a user-definable key performance indicator (KPI).

14. The system of claim 10, further comprising a graphics processing unit adapted to generate a graphical representation of the identified inter-relationship between the first and second performance metrics.

15. The system of claim 10, wherein the data analysis unit is further adapted to determine a dependency between the first and second performance metrics based on the identified inter-relationship between the first and second performance metrics,

and wherein the system is further adapted to, based on the determined dependency, monitor only one of the first and second performance metrics to detect an occurrence of an anomaly in the monitored performance metric.