Managing a storage array (102, 104) includes: receiving, by a client-side array services module (178) from a cloud-based security module (174) through data communications on a wide area network (172), a token representing authentication of user credentials; and managing, by the client-side array services module (178), a storage array (102, 104) only through data communications on a local area network (160), including sending, to the storage array (102, 104), the token with a management instruction.

Managing a storage array using client-side services

Locally providing cloud storage array services for a storage array of a data center when the storage array is not connected to a remote cloud-based storage array services provider includes initiating, by a primary storage array, one or more cloud storage array services and locally providing the cloud storage array services. Such local providing of the cloud storage array services also includes generating, by the cloud storage array services, metadata describing one or more real-time storage array characteristics and presenting the metadata to a user through a local area network.

Locally providing cloud storage array services

A pod, the pod including the dataset, a set of managed objects and management operations, a set of access operations to modify or read the dataset, and a plurality of storage systems, where: management operations can modify or query managed objects equivalently through any of the storage systems, access operations to read or modify the dataset operate equivalently through any of the storage systems, each storage system stores a separate copy of the dataset as a proper subset of the datasets stored and advertised for use by the storage system, and operations to modify managed objects or the dataset performed and completed through any one storage system are reflected in subsequent management objects to query the pod or subsequent access operations to read the dataset.

Synchronously replicating datasets and other managed objects to cloud-based storage systems

Ensuring reproducibility in an artificial intelligence infrastructure that includes one or more storage systems and one or more graphical processing unit (‘GPU’) servers, including: identifying, by a unified management plane, one or more transformations applied to a dataset by the artificial intelligence infrastructure, wherein applying the one or more transformations to the dataset causes the artificial intelligence infrastructure to generate a transformed dataset; storing, within the one or more storage systems, information describing the dataset, the one or more transformations applied to the dataset, and the transformed dataset; identifying, by the unified management plane, one or more machine learning models executed by the artificial intelligence infrastructure using the transformed dataset as input; and storing, within the one or more storage systems, information describing one or more machine learning models executed using the transformed dataset as input.

Ensuring reproducibility in an artificial intelligence infrastructure

One or more techniques and/or devices are provided for storage virtual machine relocation (eg, ownership change) between storage clusters For example, operational statistics of a first storage cluster and a second storage cluster may be evaluated to identify a set of load balancing metrics Ownership of one or more storage aggregates and/or one or more storage virtual machines may be changed (eg, permanently changed for load balancing purposes or temporarily changed for disaster recovery purposes) between the first storage cluster and the second storage cluster utilizing zero-copy ownership change operations based upon the set of load balancing metrics For example, if the first storage cluster is experiencing a relatively heavier load of client I/O operations and the second storage cluster has available resources, ownership of a storage aggregate and a storage virtual machine may be switched from the first storage cluster to the second storage cluster for load balancing

Storage virtual machine relocation

An approach is provided in which a storage system includes a first storage controller, a second storage controller, and multiple logical units. The storage system determines that a controller traffic load ratio between the first storage controller and the second storage controller has reached a threshold. In turn, the storage system selects one of the logical units and changes a preferred controller ownership of the selected logical unit from the first storage controller to the second storage controller to balance the controller traffic load ratio.

Load Balancing Logical Units in an Active/Passive Storage System

An approach is provided in which a storage system includes a first storage controller, a second storage controller, and multiple logical units. The storage system detects that a controller traffic load ratio between the first storage controller and the second storage controller has reached a threshold. As such, the storage system selects one of the logical units and partitions the selected logical unit into multiple logical units. The storage system, in turn, changes a preferred controller ownership of one of the logical unit partitions from the first storage controller to the second storage controller.

Michael S. Anderson

Papers

Load Balancing Logical Units in an Active/Passive Storage System

Balancing traffic through logical unit range partitioning