Method and device for determining a performance indicator value for predicting anomalies in a computing infrastructure from values of performance indicators

1. A method for determining a performance index value for predicting anomalies in a computing infrastructure, said method being executed by a computing device, said computing device comprising a data processing module, a storage module configured to store in memory at least one correlation base, said at least one correlation base comprising correlation data between performance indicators, and a collection module, said method comprising:

receiving, by the collection module, performance indicator values of said computing infrastructure,

identifying anomalous performance indicators, by the data processing module, said identifying anomalous performance indicators comprising an analysis of the performance indicators so as to identify abnormal values and the performance indicators associated with the abnormal values,

identifying first at-risk indicators by the data processing module from the at least one correlation base, said first at-risk indicators being said performance indicators correlated to the anomalous performance indicators,

determining, by the data processing module, an anomaly prediction performance index value comprising:

receiving new performance indicator values,

identifying new confirmed anomalous performance indicators, said identifying new confirmed anomalous performance indicators comprising an analysis of the new performance indicator values so as to identify said abnormal values and said performance indicators associated with the abnormal values, and

comparing identified at-risk indicators to the new confirmed anomalous performance indicators so as to generate said anomaly prediction performance index value for predicting anomalies in the computing infrastructure,

identifying paths that lead to a failure of the computing infrastructure for each of the performance indicators,

determining an estimated duration before incident for each of said paths,

said determining comprising a calculation, from values of duration before becoming anomalous of the identified at-risk indicators, of a shortest path leading to a risk of technical incident of said computing infrastructure,

wherein said values of duration before becoming anomalous are stored in the at least one correlation base and correspond to a duration between an occurrence of an abnormal value for a performance indicator in question and an occurrence of an abnormal value for a performance indicator correlated to the performance indicator in question,

identifying the shortest path leading to the risk of technical incident, wherein said shortest path is a quickest path that leads to said failure of said computing infrastructure,

anticipating an occurrence of said technical incident or said failure within said computing infrastructure with said shortest path within a time frame before a time of occurrence of said technical incident or said failure, and

generating a dashboard comprising

the estimated duration before incident with said shortest path to occur on said computing infrastructure or an estimated time of said technical incident to occur on said computing infrastructure, and

a percentage of confidence in said predicting; and,

facilitating intervention of a maintenance service before a service interruption is triggered and reducing an intervention time.

2. The method according to claim 1 , wherein the new performance indicator values were acquired in a course of time between an instant t 0 and an instant t n of resolution, said instant t n of resolution corresponding to an instant at which information on an incident avoided or an incident resolved is received by the computing device.

3. The method according to claim 1 , further comprising generating updated values of duration before becoming anomalous.

4. The method according to claim 3 , wherein the at least one correlation base comprises values of durations before becoming anomalous between the performance indicators, and in that the comparing of identified risk indicators with new confirmed anomaly indicators comprises a comparison between recorded values of duration before becoming anomalous between the performance indicators of the at least one correlation base and updated values of duration before becoming anomalous.

5. The method according to claim 3 , wherein the at least one correlation base comprises values of durations before becoming anomalous between the performance indicators, and the method further comprising modifying the values of durations before becoming anomalous between the performance indicators that are stored in said memory in the at least one correlation base, based on updated values of duration before becoming anomalous.

6. The method according to claim 1 , wherein the identifying anomalous performance indicators is preceded by a step of processing collected data, said step of processing the collected data comprising a removal of a normal component.

7. The method according to claim 1 , wherein the identifying of abnormal values is carried out by a statistical method making it possible to generate values of distance to normality.

8. The method according to claim 1 , wherein the identifying first at-risk indicators is carried out based on causal correlations between the performance indicators.

9. The method according to claim 1 , further comprising determining a technical incident risk value.

10. The method according to claim 1 , further comprising memorizing for each performance indicator: a unique identifier and at least one unique identifier of another performance indicator correlated to it.

11. A non-transitory computer readable medium comprising a computer program for determining an anomaly prediction performance indicator value in a computer infrastructure including instructions for executing a method for determining a performance index value for predicting anomalies in a computing infrastructure, said method being executed by a computing device, said computing device comprising a data processing module, a storage module configured to store in memory at least one correlation base, said at least one correlation base comprising correlation data between performance indicators, and a collection module, said method comprising:

receiving, by the collection module, performance indicator values of said computing infrastructure,

identifying anomalous performance indicators, by the data processing module, said identifying comprising an analysis of the performance indicators so as to identify abnormal values and performance indicators associated with the abnormal values,

identifying first at-risk indicators by the data processing module from the at least one correlation base, said first at-risk indicators being said performance indicators correlated to the anomalous performance indicators, determining, by the data processing module, an anomaly prediction performance index value comprising:

receiving new performance indicator values,

identifying new confirmed anomalous performance indicators, said identifying new confirmed anomalous performance indicators comprising an analysis of the new performance indicator values so as to identify said abnormal values and said performance indicators associated with the abnormal values, and

comparing identified at-risk indicators to the new confirmed anomalous performance indicators so as to generate said anomaly prediction performance index value for predicting anomalies in the computing infrastructure,

identifying paths that lead to a failure of the computing infrastructure for each of the performance indicators,

determining an estimated duration before incident for each of said paths,

said determining comprising a calculation, from values of duration before becoming anomalous of the identified at-risk indicators, of a shortest path leading to a risk of technical incident of said computing infrastructure,

wherein said values of duration before becoming anomalous are stored in the at least one correlation base and correspond to a duration between an occurrence of an abnormal value for a performance indicator in question and an occurrence of an abnormal value for a performance indicator correlated to the performance indicator in question,

identifying the shortest path leading to the risk of technical incident, wherein said shortest path is a quickest path that leads to said failure of said computing infrastructure,

anticipating an occurrence of said technical incident or said failure within said computing infrastructure with said shortest path within a time frame before a time of occurrence of said technical incident or said failure, and,

generating a dashboard comprising

the estimated duration before incident with said shortest path to occur on said computing infrastructure or an estimated time of said technical incident to occur on said computing infrastructure, and

a percentage of confidence in said predicting; and,

facilitating intervention of a maintenance service before a service interruption is triggered and reducing an intervention time.

12. The non-transitory computer readable medium comprising a computer program of claim 11 wherein said computer program is recorded in a storage medium.

13. A computing device for determining an anomaly prediction performance index value in a computing infrastructure, said computing device comprising a data processing module, a storage module configured to store in memory at least one correlation base comprising correlation data between performance indicators and a collection module configured to receive performance indicator values from the computer infrastructure, said data processing module being configured to:

identify anomalous performance indicators, said identifying comprising an analysis of the performance indicator values so as to identify abnormal values and performance indicators associated with the abnormal values,

identify first at-risk indicators, from the at least one correlation base, said first at-risk indicators being the performance indicators correlated to the anomalous performance indicators,

determine an anomaly prediction performance index value comprising:

receive new performance indicator values;

identify new confirmed anomalous performance indicators, said identifying new confirmed anomalous performance indicators comprising an analysis of the new performance indicator values so as to identify the abnormal values and the performance indicators associated with the abnormal values, and

compare identified at-risk indicators to the new confirmed anomalous performance indicators in order to generate a performance index value,

identify paths that lead to a failure of the computing infrastructure for each of the performance indicators,

determine an estimated duration before incident for each of said path,

said determining comprising a calculation, from values of duration before becoming anomalous of the identified at-risk indicators, of a shortest path leading to a risk of technical incident of said computing infrastructure,

wherein said values of duration before becoming anomalous are stored in the at least one correlation base and correspond to a duration between an occurrence of an abnormal value for a performance indicator in question and an occurrence of an abnormal value for a performance indicator correlated to the performance indicator in question,

identify the shortest path leading to the risk of technical incident, wherein said shortest path is a quickest path that leads to said failure of said computing infrastructure,

anticipating an occurrence of said technical incident or said failure within said computing infrastructure with said shortest path within a time frame before a time of occurrence of said technical incident or said failure, and

generate a dashboard comprising

the estimated duration before incident with said shortest path to occur on said computing infrastructure or an estimated time of said technical incident to occur on said computing infrastructure, and

a percentage of confidence in said predicting; and,

facilitating intervention of a maintenance service before a service interruption is triggered and reducing an intervention time.