Method for creating period profile for time-series data with recurrent patterns

1. A method comprising:

receiving, by a cloud service from a client of a plurality of clients of the cloud service, a request to detect and characterize seasonal patterns within a set of time series data that includes data spanning a plurality of time windows having a seasonal period;

wherein the time series data is accessible to the plurality of clients from the cloud service;

responsive to receiving the request, associating, based at least in part on the set of time series data, a first set of sub-periods of the seasonal period with a particular class of seasonal pattern;

wherein, after associating the first set of sub-periods with the particular class of seasonal pattern, a second set of sub-periods is not associated with the particular class of seasonal pattern;

generating and storing a profile for the seasonal period that identifies which sub-periods of the seasonal period are associated with the particular class of seasonal pattern;

wherein the profile is accessible to the client from the cloud service.

2. The method of claim 1 , wherein the set of time series data includes a set of historical metrics captured from one or more host devices; wherein the request identifies, locates, or provides the set of time series data to the cloud service; wherein the request comprises a request for a forecast and the profile comprises a set of forecasted values for the set of time series data; wherein the set of forecasted values is generated based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.

3. The method of claim 1 , wherein the client performs one or more seasonal-aware operations using the profile.

4. The method of claim 1 , further comprising determining whether to classify behavior of at least one computing resource as anomalous based on whether the behavior was detected in a sub-period of the seasonal period that is associated with the particular class of seasonal pattern.

5. The method of claim 1 , further comprising scheduling at least one of a maintenance operation or a batch job on at least one computing resource based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.

6. The method of claim 1 , further comprising performing at least one of consolidating or deploying computing resources based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.

7. The method of claim 1 , wherein the particular class of seasonal pattern is one of a sparse high, a dense high, a sparse low, or a dense low.

8. The method of claim 1 , wherein associating, based at least in part on the set of time series data, a first set of sub-periods of the seasonal period with a particular class of seasonal pattern comprises applying a plurality of classifiers to a plurality of sub-periods within the seasonal period; and including sub-period from the plurality of sub-periods in the first set of sub-periods that have been associated with the particular class of seasonal pattern by a threshold number of classifiers.

9. The method of claim 1 , wherein the second set of sub-periods is associated with a second class of seasonal pattern or is unclassified.

10. One or more non-transitory computer-readable media storing instructions, which, when executed by one or more hardware processors, cause performance of operations comprising:

receiving, by a cloud service from a client of a plurality of clients of the cloud service, a request to detect and characterize seasonal patterns within a set of time series data that includes data spanning a plurality of time windows having a seasonal period;

wherein the time series data is accessible to the plurality of clients from the cloud service;

responsive to receiving the request, associating, based at least in part on the set of time series data, a first set of sub-periods of the seasonal period with a particular class of seasonal pattern;

wherein, after associating the first set of sub-periods with the particular class of seasonal pattern, a second set of sub-periods is not associated with the particular class of seasonal pattern;

generating and storing a profile for the seasonal period that identifies which sub-periods of the seasonal period are associated with the particular class of seasonal pattern;

wherein the profile is accessible to the client from the cloud service.

11. The one or more non-transitory computer-readable media of claim 10 , wherein the set of time series data includes a set of historical metrics captured from one or more host devices; wherein the request identifies, locates, or provides the set of time series data to the cloud service; wherein the request comprises a request for a forecast and the profile comprises a set of forecasted values for the set of time series data; wherein the set of forecasted values is generated based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.

12. The one or more non-transitory computer-readable media of claim 10 , wherein the client performs one or more seasonal-aware operations using the profile.

13. The one or more non-transitory computer-readable media of claim 10 , the operations further comprising determining whether to classify behavior of at least one computing resource as anomalous based on whether the behavior was detected in a sub-period of the seasonal period that is associated with the particular class of seasonal pattern.

14. The one or more non-transitory computer-readable media of claim 10 , the operations further comprising scheduling at least one of a maintenance operation or a batch job on at least one computing resource based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.

15. The one or more non-transitory computer-readable media of claim 10 , the operations further comprising performing at least one of consolidating or deploying computing resources based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.

16. The one or more non-transitory computer-readable media of claim 10 , wherein the particular class of seasonal pattern is one of a sparse high, a dense high, a sparse low, or a dense low.

17. The one or more non-transitory computer-readable media of claim 10 , wherein associating, based at least in part on the set of time series data, a first set of sub-periods of the seasonal period with a particular class of seasonal pattern comprises applying a plurality of classifiers to a plurality of sub-periods within the seasonal period; and including sub-period from the plurality of sub-periods in the first set of sub-periods that have been associated with the particular class of seasonal pattern by a threshold number of classifiers.

18. The one or more non-transitory computer-readable media of claim 10 , wherein the second set of sub-periods is associated with a second class of seasonal pattern or is unclassified.

19. A system comprising:

one or more hardware processors;

one or more non-transitory computer-readable media storing instructions, which, when executed by one or more hardware processors, cause performance of operations comprising:

receiving, by a cloud service from a client of a plurality of clients of the cloud service, a request to detect and characterize seasonal patterns within a set of time series data that includes data spanning a plurality of time windows having a seasonal period;

wherein the time series data is accessible to the plurality of clients from the cloud service;

responsive to receiving the request, associating, based at least in part on the set of time series data, a first set of sub-periods of the seasonal period with a particular class of seasonal pattern;

wherein, after associating the first set of sub-periods with the particular class of seasonal pattern, a second set of sub-periods is not associated with the particular class of seasonal pattern;

generating and storing a profile for the seasonal period that identifies which sub-periods of the seasonal period are associated with the particular class of seasonal pattern;

wherein the profile is accessible to the client from the cloud service.

20. The system of claim 19 , wherein the set of time series data includes a set of historical metrics captured from one or more host devices; wherein the request identifies, locates, or provides the set of time series data to the cloud service; wherein the request comprises a request for a forecast and the profile comprises a set of forecasted values for the set of time series data; wherein the set of forecasted values is generated based on which sub-periods of the seasonal period are associated with the particular class of seasonal pattern.