Classification of storage systems and users thereof using machine learning techniques
Methods, apparatus, and processor-readable storage media for classification of storage systems and users thereof using machine learning techniques are provided herein. An example computer-implemented method includes processing input data pertaining to multiple storage systems within an enterprise; classifying one or more of the storage systems by applying a first set of machine learning techniques to the processed input data; classifying one or more respective users of the classified storage systems by applying a second set of machine learning techniques to the processed input data associated with the classified storage systems; and outputting, via one or more user interfaces, at least a portion of the storage system classifications and at least a portion of the user classifications to a user for use in connection with storage system configuration actions and/or an entity within the enterprise for use in connection with user-support actions.
1. A computer-implemented method comprising:
processing input data pertaining to multiple storage systems within an enterprise, wherein processing the input data comprises performing multiple feature engineering steps on the input data, and wherein performing the multiple feature engineering steps comprises cleaning the input data, imputing one or more items of missing data in the input data, and converting, in the input data, one or more items of numerical data to one or more items of categorical data;
classifying one or more of the multiple storage systems by applying a first set of one or more machine learning techniques to the processed input data;
classifying one or more respective users of the one or more classified storage systems by applying a second set of one or more machine learning techniques to at least a portion of the processed input data associated with the one or more classified storage systems; and
outputting, via one or more user interfaces, at least a portion of the storage system classifications and at least a portion of the user classifications to one or more of a user for use in connection with one or more storage system configuration actions and an entity within the enterprise for use in connection with one or more user-support actions;
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.
2. The computer-implemented method of claim 1 , wherein the first set of one or more machine learning techniques comprises at least one unsupervised k-means machine learning clustering algorithm that analyzes at least a portion of the processed input data against a feature dataset, and ranks the multiple storage systems based at least in part of the analysis.
3. The computer-implemented method of claim 2 , wherein the second set of one or more machine learning techniques comprises at least one unsupervised k-means machine learning clustering algorithm that processes at least a portion of the processed input data against the raking of the multiple storage systems and the feature dataset, and generates a rating for each of the users based at least in part of the processing.
4. The computer-implemented method of claim 1 , wherein processing the input data comprises processing a portion of the input data encompassing a given temporal period.
5. The computer-implemented method of claim 1 , wherein classifying the one or more storage systems comprises classifying the one or more storage systems into one of multiple predetermined storage system classification groups.
6. The computer-implemented method of claim 1 , wherein classifying the one or more respective users comprises classifying the one or more respective users into one of multiple predetermined user classification groups.
7. The computer-implemented method of claim 1 , wherein the input data comprise one or more of storage system configuration data associated with the multiple storage systems, storage system operations data associated with the multiple storage systems, storage system heuristic-based health scores associated with the multiple storage systems, and user service request counts associated with the multiple storage systems.
8. The computer-implemented method of claim 1 , further comprising:
automatically configuring one or more of the storage systems by executing the one or more storage system configuration actions.
9. The computer-implemented method of claim 1 , further comprising:
automatically performing the one or more user-support actions by updating a data structure associated with at least one user.
10. The computer-implemented method of claim 1 , wherein outputting the at least a portion of the storage system classifications comprises generating, via the one or more user interfaces, a view representing one or more performance levels for a given one of the storage systems relative to all of the storage systems within the enterprise.
11. The computer-implemented method of claim 1 , wherein outputting the at least a portion of the user classifications comprises generating, via the one or more user interfaces, a view representing one or more performance levels for a given one of the users relative to all of the users within the enterprise.
12. The computer-implemented method of claim 1 , further comprising:
generating a comparison between at least one of the classified users and one or more additional classified users by applying one or more heuristic calculations to at least a portion of the processed input data associated with the classified users; and
presenting, via at least one of the one or more user interfaces, the comparison to the at least one classified user.
13. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:
to process input data pertaining to multiple storage systems within an enterprise, wherein processing the input data comprises performing multiple feature engineering steps on the input data, and wherein performing the multiple feature engineering steps comprises cleaning the input data, imputing one or more items of missing data in the input data, and converting, in the input data, one or more items of numerical data to one or more items of categorical data;
to classify one or more of the multiple storage systems by applying a first set of one or more machine learning techniques to the processed input data;
to classify one or more respective users of the one or more classified storage systems by applying a second set of one or more machine learning techniques to at least a portion of the processed input data associated with the one or more classified storage systems; and
to output, via one or more user interfaces, at least a portion of the storage system classifications and at least a portion of the user classifications to one or more of a user for use in connection with one or more storage system configuration actions and an entity within the enterprise for use in connection with one or more user-support actions.
14. The non-transitory processor-readable storage medium of claim 13 , wherein the program code further causes the at least one processing device:
to automatically configure one or more of the storage systems by executing the one or more storage system configuration actions.
15. The non-transitory processor-readable storage medium of claim 13 , wherein the program code further causes the at least one processing device:
to automatically perform the one or more user-support actions by updating a data structure associated with at least one user.
16. An apparatus comprising:
at least one processing device comprising a processor coupled to a memory;
the at least one processing device being configured:
to process input data pertaining to multiple storage systems within an enterprise, wherein processing the input data comprises performing multiple feature engineering steps on the input data, and wherein performing the multiple feature engineering steps comprises cleaning the input data, imputing one or more items of missing data in the input data, and converting, in the input data, one or more items of numerical data to one or more items of categorical data;
to classify one or more of the multiple storage systems by applying a first set of one or more machine learning techniques to the processed input data;
to classify one or more respective users of the one or more classified storage systems by applying a second set of one or more machine learning techniques to at least a portion of the processed input data associated with the one or more classified storage systems; and
to output, via one or more user interfaces, at least a portion of the storage system classifications and at least a portion of the user classifications to one or more of a user for use in connection with one or more storage system configuration actions and an entity within the enterprise for use in connection with one or more user-support actions.
17. The apparatus of claim 16 , wherein the at least one processing device is further configured:
to automatically configure one or more of the storage systems by executing the one or more storage system configuration actions.
18. The apparatus of claim 16 , wherein the at least one processing device is further configured:
to automatically perform the one or more user-support actions by updating a data structure associated with at least one user.