Abstract:
In some examples, a fleet storage provider performs storage management operations for a fleet of storage arrays, the storage arrays in the fleet of storage arrays being of one or more storage types. In response to an addition, to the fleet of storage arrays, of a new storage array of a first storage type different from each storage type of the one or more storage types, a system identifies the new storage array as being associated with a first storage class of a plurality of different storage classes, and associates the new storage array with a fleet service that abstracts component details of the fleet of storage arrays to the fleet storage provider. In response to a request, the system provisions a storage volume on a selected storage array of the fleet of storage arrays, the provisioning performed by the fleet storage provider in cooperation with the fleet service.
Abstract:
In some examples, a fleet storage provider performs storage management operations for a fleet of storage arrays, the storage arrays in the fleet of storage arrays being of one or more storage types. In response to an addition, to the fleet of storage arrays, of a new storage array of a first storage type different from each storage type of the one or more storage types, a system identifies the new storage array as being associated with a first storage class of a plurality of different storage classes, and associates the new storage array with a fleet service that abstracts component details of the fleet of storage arrays to the fleet storage provider. In response to a request, the system provisions a storage volume on a selected storage array of the fleet of storage arrays, the provisioning performed by the fleet storage provider in cooperation with the fleet service.
Abstract:
In some examples, a system creates a training data set based on features of sample workloads, the training data set comprising labels associated with the features of the sample workloads, where the labels are based on load indicators generated in a computing environment relating to load conditions of the computing environment resulting from execution of the sample workloads. The system groups selected workloads into a plurality of workload clusters based on features of the selected workloads, and computes, using a model trained based on the training data set, parameters representing contributions of respective workload clusters of the plurality of workload clusters to a load in the computing environment. The system performs workload management in the computing environment based on the computed parameters.
Abstract:
In some examples, a system receives input information of characteristics relating to a storage volume to be provisioned in a collection of storage systems, determines, based on the input information of the characteristics relating to the storage volume, a workload profile, and simulates execution of a workload according to the workload profile in each storage system of the collection of storage systems. Based on the simulation, the system determines a respective amount of headroom used by the workload in each storage system of the collection of storage systems, and selects, based on the determined respective amounts of headroom used by the workload in respective storage systems of the collection of storage systems, a storage system from the collection of storage systems on which the storage volume is to be provisioned.
Abstract:
In some examples, using a model generated from an aggregation of parameter values for a plurality of host systems, a system predicts an operational metric representing usage or performance of a shared resource due to a requester in a first host system of the plurality of host systems, the shared resource being outside of the plurality of host systems.
Abstract:
In some examples, a system creates a training data set based on features of sample workloads, the training data set comprising labels associated with the features of the sample workloads, where the labels are based on load indicators generated in a computing environment relating to load conditions of the computing environment resulting from execution of the sample workloads. The system groups selected workloads into a plurality of workload clusters based on features of the selected workloads, and computes, using a model trained based on the training data set, parameters representing contributions of respective workload clusters of the plurality of workload clusters to a load in the computing environment. The system performs workload management in the computing environment based on the computed parameters.
Abstract:
In some examples, a system creates a training data set based on features of sample workloads, the training data set comprising labels associated with the features of the sample workloads, where the labels are based on load indicators generated in a computing environment relating to load conditions of the computing environment resulting from execution of the sample workloads. The system groups selected workloads into a plurality of workload clusters based on features of the selected workloads, and computes, using a model trained based on the training data set, parameters representing contributions of respective workload clusters of the plurality of workload clusters to a load in the computing environment. The system performs workload management in the computing environment based on the computed parameters.
Abstract:
The present subject matter relates to perform proactive monitoring and diagnostics in storage area networks (SANs). In one implementation, the method comprises depicting topology of the SAN in a graph, wherein the graph designates the devices as nodes, the connecting elements as edges, and depicts operations associated with at least one component of the nodes and edges. The method further comprises monitoring at least one parameter indicative of performance of the component to ascertain degradation of the at least one component and identifying, a hinge in the data associated with the monitoring, wherein the hinge is indicative of an initiation in degradation of the component. Based on the hinge, proactive diagnostics is preformed to compute a remaining lifetime of the at least one component. Thereafter, a notification is generated for an administrator of the SAN based on the remaining lifetime.