| 摘要 |
Integrated computing systems with independently managed infrastructures including compute nodes and storage nodes form a cluster. Storage resource agents manage storage resources in the cluster. The resource agents identify storage requirements associated with allocation sets for resource consumers dispatched in the cluster, communicate with each other to locate inter-system storage resources that primarily satisfy locality criteria associated with resource consumer workloads, secondarily satisfy allocation set activity criteria associated with the allocation sets, and allocate the storage resources to the resource consumers to satisfy the storage requirements. The storage resource agents may base storage assignments on data placement information from a priority map. Data may be later relocated to alternate storage resources in satisfaction of cluster-wide storage policies, priority determinations, and data access rate determinations. |
| 主权项 |
1. A method for managing data storage resources in a cluster of integrated systems, each integrated system having an integrated resource infrastructure, the integrated resource infrastructure including an at least one compute node and intra-system storage, the intra-system storage including data storage resources in an at least one storage node, the integrated resource infrastructure of each integrated system managed independently of the integrated resource infrastructure of each of the other integrated systems in the cluster, the method comprising:
associating a storage resource agent with each integrated system, each storage resource agent included in a plurality of distributed storage resource agents, the plurality of distributed storage resource agents communicatively interconnected and managing inter-system storage, the inter-system storage including the data storage resources; forming, by at least two storage resource agents associated with the integrated systems, an inter-system storage priority map, the contents of the inter-system storage priority map synchronized by the at least two storage resource agents, the inter-system storage priority map including data set allocation criteria associated with each data allocation set, the data set allocation criteria associated with each data allocation set including at least one of a storage directive, an inter-system storage priority for accessing data storage resources in the inter-system storage, and data allocation set attributes descriptive of the data allocation set; identifying, by a first storage resource agent associated with a first integrated system in the cluster, a first data storage requirement for a first resource consumer, the first data storage requirement associated with a first data allocation set, the first data allocation set associated with a workload of the first resource consumer, the first data allocation set based at least in part on data associated with the first data storage requirement, the first resource consumer dispatched on a first virtual machine hosted on a first compute node included in the first integrated system; locating, by the first storage resource agent and an at least one second storage resource agent associated with an integrated system that is not the first integrated system, an at least one candidate data storage resource in the inter-system storage, the at least one candidate data storage resource within the data storage resources of the intra-system storage of at least one integrated system within the cluster; selecting a data storage resource from the at least one candidate data storage resource, the selecting utilizing first data set allocation criteria included in the inter-system storage priority map, the first data set allocation criteria associated with the first data allocation set, the selected data storage resource satisfying a data allocation criterion included in the first data set allocation criteria; and allocating the selected data storage resource to the first resource consumer to satisfy the first data storage requirement. |
