| 摘要 |
Operational characteristic I/O patterns of each storage volume of a storage volume group, and at least one operational characteristic of each storage volume associated with an application, are determined. Unsupervised learning is used to form clusters of storage volumes of similar characteristics. Labels are generated, assigned, and aggregated for each characteristic of each storage volume. Templates are received that includes labels of storage volume characteristics of known application types. An application type associated with the storage volume group is inferred, based on a best-fit match of the aggregate labels of the storage volumes of the storage volume group to the labels included in the templates of storage volume characteristics of known application types. |
| 主权项 |
1. A method for inferring a type of an application to which a storage volume is associated, the method comprising:
determining, by one or more processors, at least one characteristic of a plurality of characteristics of storage volume operational metrics for each storage volume of a storage volume group associated with an application, wherein the at least one characteristic includes a pattern of input-output (I/O) of a storage volume; creating, by one or more processors, a plurality of clusters of storage volumes by utilizing unsupervised machine learning techniques to the at least one characteristic of the plurality of characteristics of storage volume operational metrics, wherein a first cluster of the plurality of clusters includes storage volumes that have a similar attribute of the at least one characteristic that includes a pattern of input-output, and excludes storage volumes that lack the similar attribute of the at least one characteristic that includes the pattern of input-output; generating, by one or more processors, a label for each of the storage volumes of the first cluster of the plurality of clusters, which corresponds to the similar attribute of the at least one characteristic of the plurality of characteristics of each storage volume of the first cluster, and generating a label for each of the storage volumes of each cluster of the plurality of clusters, such that a particular label corresponds to storage volumes of a particular cluster having an attribute of a particular characteristic of the at least one characteristic that includes a pattern of input-output; assigning, by one or more processors, one or more labels to each storage volume of the storage volume group associated with the application, wherein each label of the one or more labels corresponds to a cluster of the plurality of clusters, and each storage volume of the storage volume group is included in one or more clusters by the unsupervised machine learning determining that a storage volume has an attribute of the at least one characteristic corresponding to a cluster of the one or more clusters; receiving, by one or more processors, at least one template of storage volume characteristics of known application types, wherein each template includes a set of labels that are based on the storage volume characteristics associated with a known application type; and inferring, by one or more processors, a type of application associated with the storage volume group, based on a best-fit match of a combination of the one or more labels of each storage volume of the storage volume group to a set of labels that are associated with the at least one template of storage volume characteristics of known application types. |
