Anomaly detection based on cluster transitions

1. A method comprising:

receiving first performance data of a plurality of components in a system, the first performance data representing performance characteristics of the plurality of components in a first time period;

performing a first cluster analysis of the first performance data, the first cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the first time period;

receiving second performance data of the plurality of components, the second performance data representing performance characteristics of the plurality of components in a second time period;

performing a second cluster analysis of the second performance data, the second cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the second time period;

determining whether a component of the plurality of components transitioned from a first cluster to a second cluster, the first cluster being different from the second cluster, and the determining comprising:

determining that the component transitioned from the first cluster to the second cluster if the component was identified as being in the first cluster in the first cluster analysis and was identified as being in the second cluster in the second cluster analysis; and

determining that the component did not transition from the first cluster to the second cluster if the component was identified as being in the first cluster in the first cluster analysis and was identified as being in the first cluster in the second cluster analysis; and

in response to determining that the component transitioned from the first cluster to the second cluster, determining that an anomalous event has occurred;

receiving third performance data of the plurality of components, the third performance data representing performance characteristics of the plurality of components in a third time period;

performing a third cluster analysis of the third performance data, the third cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics representing in the third time period;

determining whether the component transitioned from the second cluster to a third cluster, the third cluster being different from the second cluster, and the determining comprising:

determining that the component transitioned from the second cluster to the third cluster if the component was identified as being in the second cluster in the second cluster analysis and was identified as being in the third cluster in the third cluster analysis; and

in response to determining that the component transitioned from the second cluster to the third cluster:

determining that the component is in a steady state of transition, and

determining whether the component is a member of a transition cluster that is in a steady state of transition between the first cluster and the third cluster.

2. The method of claim 1 ,

wherein performing the first cluster analysis comprises clustering the plurality of components at a first level of granularity,

wherein performing the second cluster analysis comprises clustering the plurality of components at a second level of granularity, and

wherein the first level of granularity is the same level of granularity as the second level of granularity, such that the first cluster and the second cluster are clusters at the same level of granularity.

3. The method of claim 1 ,

wherein performing the first cluster analysis comprises:

determining that the component has similar performance characteristics to another component of the plurality of components based on the first performance data; and

in response to determining that the component has similar performance characteristics to the other component, determining that the component and the other component are members of the first cluster, and

wherein performing the second cluster analysis comprises:

determining whether the component has similar performance characteristics to the other component based on the second performance data; and

in response to determining that the component has similar performance characteristics to the other component, determining that the component and the other component are members of the first cluster; and

in response to determining that the component does not have similar performance characteristics to the other component, determining that the component is a member of the second cluster.

4. The method of claim 1 ,

wherein performing the first cluster analysis comprises clustering the plurality of components at each of a plurality of levels of granularity,

wherein performing the second cluster analysis comprises clustering the plurality of components at each of the plurality of levels of granularity, and

wherein determining whether the component transitioned from the first cluster to the second cluster comprises determining whether the component transitioned from the first cluster to the second cluster based on clustering the plurality of components at the same level of granularity in both the first cluster analysis and the second cluster analysis.

5. The method of claim 1 ,

wherein performing the first cluster analysis of the first performance data comprises:

identifying a cluster of components of the plurality of components with similar performance characteristics for a first parameter of the first performance data in the first time period, such cluster being a first parameter cluster; and

identifying a cluster of components of the plurality of components with similar performance characteristics for a second parameter of the first performance data in the first time period, such cluster being the first cluster,

wherein the method further comprises:

requesting data for a particular parameter of performance data from the components identified as the first parameter cluster, the particular parameter being the same parameter as the second parameter of the first performance data,

wherein receiving the second performance data of the plurality of components comprises:

receiving as the second performance data the data for the particular parameter from the components identified as the first parameter cluster, and

wherein performing the second cluster analysis of the second performance data comprises:

identifying a plurality of clusters of components of the plurality of components with similar performance characteristics for the particular parameter of performance data in the second time period, such plurality of clusters comprising the first cluster and the second cluster.

6. The method of claim 1 , further comprising:

accessing additional information, the additional information indicating that transitioning from the first cluster to the second cluster is acceptable behavior for the component of the plurality of components, and

in response to accessing the additional information, determining that the anomalous event is an acceptable event.

7. A system comprising:

a memory; and

a processing system configured to:

receive first performance data of a plurality of components in a system, the first performance data performance characteristics of the plurality of components in a first time period;

perform a first cluster analysis of the first performance data, the first cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the first time period;

receive second performance data of the plurality of components, the second performance data representing performance characteristics of the plurality of components in a second time period;

perform a second cluster analysis of the second performance data, the second cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the second time period;

determine whether a component of the plurality of components transitioned from a first cluster to a second cluster, the first cluster being different from the second cluster, and the determining comprising:

determining that the component transitioned from the first cluster to the second cluster if the component was identified as being in the first cluster in the first cluster analysis and was identified as being in the second cluster in the second cluster analysis; and

determining that the component did not transition from the first cluster to the second cluster if the component was identified as being in the first cluster in the first cluster analysis and was identified as being in the first cluster in the second cluster analysis; and

in response to determining that the component transitioned from the first cluster to the second cluster, determine that an anomalous event has occurred;

receiving third performance data of the plurality of components, the third performance data representing performance characteristics of the plurality of components in a third time period;

performing a third cluster analysis of the third performance data, the third cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the third time period;

determining whether the component transitioned from the second cluster to a third cluster, the third cluster being different from the second cluster, and the determining comprising:

determining that the component transitioned from the second cluster to the third cluster if the component was identified as being in the second cluster in the second cluster analysis and was identified as being in the third cluster in the third cluster analysis; and

in response to determining that the component transitioned from the second cluster to the third cluster:

determining that the component is in a steady state of transition, and

determining whether the component is a member of a transition cluster that is in a steady state of transition between the first cluster and the third cluster.

8. The system according to claim 7 ,

wherein, when performing the first cluster analysis, the processing system is configured to cluster the plurality of components at a first level of granularity,

wherein, when performing the second cluster analysis, the processing system is configured to cluster the plurality of components at a second level of granularity, and

wherein the first level of granularity is the same level of granularity as the second level of granularity, such that the first cluster and the second cluster are clusters at the same level of granularity.

9. The system according to claim 7 ,

wherein, when performing the first cluster analysis, the processing system is configured to:

determine that the component has similar performance characteristics to another component of the plurality of components based on the first performance data; and

in response to determining that the component has similar performance characteristics to the other component, determine that the component and the other component are members of the first cluster, and

wherein, when performing the second cluster analysis, the processing system is configured to:

determine whether the component has similar performance characteristics to the other component based on the second performance data; and

in response to determining that the component has similar performance characteristics to the other component, determine that the component and the other component are members of the first cluster; and

in response to determining that the component does not have similar performance characteristics to the other component, determine that the component is a member of the second cluster.

10. The system according to claim 7 ,

wherein, when performing the first cluster analysis, the processing system is configured to cluster the plurality of components at each of a plurality of levels of granularity,

wherein, when performing the second cluster analysis, the processing system is configured to cluster the plurality of components at each of the plurality of levels of granularity, and

wherein, when determining whether the component transitioned from the first cluster to the second cluster, the processing system is configured to determine whether the component transitioned from the first cluster to the second cluster based on clustering the plurality of components at the same level of granularity in both the first cluster analysis and the second cluster analysis.

11. The system according to claim 7 ,

wherein, when performing the first cluster analysis of the first performance data, the processing system is configured to:

identify a cluster of components of the plurality of components with similar performance characteristics for a first parameter of the first performance data in the first time period, such cluster being a first parameter cluster; and

identify a cluster of components of the plurality of components with similar performance characteristics for a second parameter of the first performance data in the first time period, such cluster being the first cluster,

wherein the processing system is further configured to:

request data for a particular parameter of performance data from the components identified as the first parameter cluster, the particular parameter being the same parameter as the second parameter of the first performance data,

wherein, when receiving the second performance data of the plurality of components, the processing system is configured to:

receive as the second performance data the data for the particular parameter from the components identified as the first parameter cluster, and

wherein, when performing the second cluster analysis of the second performance data, the processing system is configured to:

identify a plurality of clusters of components of the plurality of components with similar performance characteristics for the particular parameter of performance data in the second time period, such plurality of clusters comprising the first cluster and the second cluster.

12. The system according to claim 7 , wherein the processing system is further configured to:

access additional information, the additional information indicating that transitioning from the first cluster to the second cluster is acceptable behavior for the component of the plurality of components, and

in response to accessing the additional information, determine that the anomalous event is an acceptable event.

13. A computer program product comprising:

a non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising:

computer readable program code configured to receive first performance data of a plurality of components in a system, the first performance data representing performance characteristics of the plurality of components in a first time period;

computer readable program code configured to perform a first cluster analysis of the first performance data, the first cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the first time period;

computer readable program code configured to receive second performance data of the plurality of components, the second performance data representing performance characteristics of the plurality of components in a second time period;

computer readable program code configured to perform a second cluster analysis of the second performance data, the first cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the second time period;

computer readable program code configured to determine whether a component of the plurality of components transitioned from a first cluster to a second cluster, the first cluster being different from the second cluster, and the determining comprising:

determining that the component transitioned from the first cluster to the second cluster if the component was identified as being in the first cluster in the first cluster analysis and was identified as being in the second cluster in the second cluster analysis; and

determining that the component did not transition from the first cluster to the second cluster if the component was identified as being in the first cluster in the first cluster analysis and was identified as being in the first cluster in the second cluster analysis; and

computer readable program code configured to, in response to determining that the component transitioned from the first cluster to the second cluster, determine that an anomalous event has occurred;

computer readable program code configured to receive third performance data of the plurality of components, the third performance data representing performance characteristics of the plurality of components in a third time period;

computer readable program code configured to perform a third cluster analysis of the third performance data, the third cluster analysis comprising:

identifying clusters of the plurality of components with similar performance characteristics in the third time period;

computer readable program code configured to determine whether the component transitioned from the second cluster to a third cluster, the third cluster being different from the second cluster, and the determining comprising:

determining that the component transitioned from the second cluster to the third cluster if the component was identified as being in the second cluster in the second cluster analysis and was identified as being in the third cluster in the third cluster analysis; and

computer readable program code configured to, in response to determining that the component transitioned from the second cluster to the third cluster:

determine that the component is in a steady state of transition; and

determine whether the component is a member of a transition cluster that is in a steady state of transition between the first cluster and the third cluster.

14. The computer program product of claim 13 ,

wherein the computer readable program code configured to perform the first cluster analysis comprises:

computer readable program code configured to cluster the plurality of components at a first level of granularity,

wherein the computer readable program code configured to perform the second cluster analysis comprises:

computer readable program code configured to cluster the plurality of components at a second level of granularity, and

wherein the first level of granularity is the same level of granularity as the second level of granularity, such that the first cluster and the second cluster are clusters at the same level of granularity.

15. The computer program product of claim 13 ,

wherein the computer readable program code configured to perform the first cluster analysis comprises:

determining that the component has similar performance characteristics to another component of the plurality of components based on the first performance data; and

in response to determining that the component has similar performance characteristics to the other component, determining that the component and the other component are members of the first cluster, and

wherein the computer readable program code configured to perform the second cluster analysis comprises:

computer readable program code configured to determine whether the component has similar performance characteristics to the other component based on the second performance data; and

computer readable program code configured to, in response to determining that the component has similar performance characteristics to the other component, determine that the component and the other component are members of the first cluster; and

computer readable program code configured to, in response to determining that the component does not have similar performance characteristics to the other component, determine that the component is a member of the second cluster.

16. The computer program product of claim 13 ,

wherein the computer readable program code configured to perform the first cluster analysis comprises:

computer readable program code configured to cluster the plurality of components at each of a plurality of levels of granularity,

wherein the computer readable program code configured to perform the second cluster analysis comprises:

computer readable program code configured to cluster the plurality of components at each of the plurality of levels of granularity, and

wherein the computer readable program code configured to determine whether the component transitioned from the first cluster to the second cluster comprises:

computer readable program code configured to determine whether the component transitioned from the first cluster to the second cluster based on clustering the plurality of components at the same level of granularity in both the first cluster analysis and the second cluster analysis.

17. The computer program product of claim 13 , wherein the computer program product further comprises:

computer readable program code configured to access additional information, the additional information indicating that transitioning from the first cluster to the second cluster is acceptable behavior for the component of the plurality of components, and

computer readable program code configured to, in response to accessing the additional information, determine that the anomalous event is an acceptable event.