Facilitating detection of anomalies in data center telemetry
Facilitating detection of anomalies of a target entity is provided herein. A system can comprise a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations. The operations can comprise training a model on a first set of variables that are constrained by a second set of variables. The second set of variables can characterize elements of a defined entity. The first set of variables can define a normality of the defined entity. The operations also can comprise employing the model to identify expected parameters and unexpected parameters associated with the defined entity to at least a defined level of confidence.
1. A method, comprising:
training, by a system comprising a processor, a model on a first set of variables that are constrained by a second set of variables, wherein the second set of variables characterizes elements of a defined entity, and wherein the first set of variables defines a normality and an anomaly of the defined entity; and
employing, by the system, the model to identify a normal state or an anomalous state of the defined entity to at least a defined level of confidence;
transforming, by the system, the second set of variables associated with the defined entity into a derivative representation that comprises the first set of variables for the defined entity, wherein the transforming comprises grouping elements of the defined entity based on a similarity function, wherein the similarity function is based on a pairwise similarity between the elements of the defined entity, and wherein the pairwise similarity is based on combining, by the system, variables of different natures of the second set of variables into a single weighted similarity function.
2. The method of claim 1 , further comprising:
detecting, by the system, an anomaly of the defined entity based on the first set of variables, wherein the second set of variables provides context for a definition of normality and a detection of anomalies associated with the defined entity.
3. The method of claim 2 , further comprising:
prior to the detecting the anomaly, constraining, by the system, the first set of variables based on the derivative representation of the second set of variables.
4. The method of claim 1 , wherein the variables of different natures comprise configuration variables of the elements of the defined entity and conditions of usage variables of the elements of the defined entity.
5. The method of claim 1 , further comprising:
assessing, by the system, the first set of variables associated with normal behavior to a group to which the first set of variables belong; and
classifying, by the system, a test sample as normal or anomalous based on the assessing.
6. The method of claim 5 , wherein the first set of variables comprise performance variables.
7. The method of claim 1 , wherein the defined entity is a storage system, and wherein the first set of variables and the second set of variables are telemetry data variables of the storage system.
8. A system, comprising:
a processor; and
a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising:
training a model on a first set of variables that are constrained by a second set of variables, wherein the second set of variables characterize elements of a defined entity, and wherein the first set of variables define a normality of the defined entity;
employing the model to identify expected parameters and unexpected parameters associated with the defined entity to at least a defined level of confidence;
transforming the second set of variables into a representation that comprises the first set of variables; and
combining variables of different natures of the second set of variables into a single weighted similarity function, resulting in a pairwise similarity, wherein the similarity function is based on the pairwise similarity between elements of the defined entity, and wherein the variables of different natures comprise configuration variables of the elements of the defined entity and conditions of usage variables of the elements of the defined entity.
9. The system of claim 8 , wherein the operations further comprise:
performing anomaly detection of the defined entity based on the first set of variables, wherein the first set of variables provide context for a definition of normality and a detection of anomalies associated with the defined entity.
10. The system of claim 9 , wherein the transforming the second set of variables comprises constraining the first set of variables based on a derivative representation of the second set of variables.
11. The system of claim 10 , wherein the operations further comprise:
evaluating performance variables associated with normal behavior of the defined entity to a group to which the second set of variables belong; and
classifying a test sample as normal or anomalous based on result of the evaluating.
12. The system of claim 10 , wherein the transforming the second set of variables comprises grouping elements of the defined entity based on a similarity function.
13. The system of claim 8 , wherein the defined entity is a data center, and wherein the first set of variables and the second set of variables are telemetry data variables of the data center.
14. A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
training a model on a first group of variables that are constrained by a second group of variables, wherein the second group of variables characterizes elements of a defined entity, and wherein the first group of variables defines a normality and an anomaly of the defined entity; and
employing the model to identify a normal state or an anomalous state of the defined entity to a defined level of confidence;
transforming the second group of variables associated with the defined entity into a derivative representation that comprises the first group of variables for the defined entity; and
grouping elements of the defined entity based on a similarity function, wherein the similarity function is based on a pairwise similarity between elements of the defined entity, wherein the pairwise similarity is based on a combination of variables of different natures of the second group of variables into a single weighted similarity function.
15. The non-transitory machine-readable medium of claim 14 , wherein the operations further comprise:
detecting an anomaly of the defined entity based on the first group of variables, wherein the second group of variables provides context for a definition of normality and a detection of anomalies associated with the defined entity.
16. The non-transitory machine-readable medium of claim 15 , wherein the operations further comprise constraining the first group of variables based on the derivative representation of the second group of variables.
17. The non-transitory machine-readable medium of claim 14 , wherein the variables of different natures comprise configuration variables of the elements of the defined entity and conditions of usage variables of the elements of the defined entity.
18. The non-transitory machine-readable medium of claim 14 , wherein the operations further comprise:
assessing the first group of variables associated with the normality to a group to which the first group of variables belong; and
classifying a test sample as normal or anomalous based on the assessing.
19. The non-transitory machine-readable medium of claim 14 , wherein the first group of variables comprise performance variables.
20. The non-transitory machine-readable medium of claim 14 , wherein the defined entity is a distributed storage system, and wherein the first group of variables and the second group of variables are telemetry data variables of the distributed storage system.