International Business Machines Corporation

Taylor D Peoples

Joseph W Cropper

Jennifer D Mulsow

Jeffrey Wayne Tenner

Sadek Jbara

Edward Shvartsman

Andrew Thomas Thorstensen

Asset management for a shared pool of configurable computing resources is disclosed. For a set of assets which share a resource on a first compute node, a set of asset weight values is identified. The set of asset weight values indicates a set of utilization shares of the resource by the set of assets. A set of asset priority values is identified with respect to the set of assets on the first compute node. The set of asset priority values indicates a relative status arrangement for the set of assets. Using the set of asset weight values and the set of asset priority values, a determination is made to migrate a first asset of the set of assets from the first compute node to a second compute node. Accordingly, the first asset of the set of assets is migrated from the first compute node to the second compute node.

Asset management with respect to a shared pool of configurable computing resources

Disclosed aspects relate to asset management with respect to a shared pool of configurable computing resources. With respect to a set of assets which share a resource on a first compute node, a set of asset weight values is identified. The set of asset weight values indicates a set of utilization shares of the resource by the set of assets. A set of asset priority values is identified with respect to the set of assets on the first compute node. The set of asset priority values indicates a relative status arrangement for the set of assets. Using the set of asset weight values and the set of asset priority values, a determination is made to migrate a first asset of the set of assets from the first compute node to a second compute node. Accordingly, the first asset of the set of assets is migrated from the first compute node to the second compute node.

Disclosed aspects manage a shared pool of configurable computing resources. A resource demand value is identified for a target host. Resource supply values of dynamically-assigned resources are computed for donor hosts. Using the resource demand value and the resource supply values, dynamically-assigned resources are reclaimed from the donor hosts and distributed to the target host.

Dynamically-assigned resource management in a shared pool of configurable computing resources

Dynamically setting the order of optimization of physical hosts allows more efficient and varied optimization. An ordering policy mechanism utilizes ordering policies to set an order for the optimizer to optimize physical the hosts. The ordering policy mechanism may allow a system administrator to create and/or select ordering policies. The ordering policies may include fixed ordering policies or dynamic ordering policies.

Ordering optimization of host machines in a computing environment based on policies

Disclosed aspects relate to managing link aggregation with respect to a shared pool of configurable computing resources. A set of workloads of the shared pool of configurable computing resources can be monitored to identify a set of networking data. Based on the set of networking data, a link aggregation arrangement for dynamic adjustment may be determined. The link aggregation arrangement may be established by dynamic adjustment with respect to the shared pool of configurable computing resources.

Link aggregation management with respect to a shared pool of configurable computing resources

Systems, methods, and computer program products configured to perform an operation comprising determining that use of a system resource of a first host system exceeds a threshold, wherein the first host system executes a plurality of virtual machines in a cloud computing environment, identifying a first virtual machine, of the plurality of virtual machines, for migration based on a first migration policy, of a plurality of migration policies, wherein the plurality of migration policies specify a respective set of rules for identifying virtual machines for migration, and migrating the first virtual machine to a second host system in the cloud computing environment to reduce the use of the system resource on the first host system.

Policy based virtual machine selection during an optimization cycle

Dynamically changing the aggressiveness of optimization of virtual machines on physical hosts allows more efficient and varied optimization. An aggressiveness policy mechanism periodically applies system conditions to the aggressiveness policies to create aggressiveness settings that are provided to an optimizer. The optimizer then uses the aggressiveness settings to dynamically adjust the aggressiveness of placement of virtual machines according to the aggressiveness settings and consistent with other optimization policies. The aggressiveness policy mechanism may allow a system administrator to create and/or select aggressiveness policies.

Dynamic aggressiveness for optimizing placement of virtual machines in a computing environment

Placement of virtual machines on physical hosts is based on a preferred host list that may be generated from preferred host identifiers to more efficiently utilize system resources and serve the needs of different workloads. A host identifiers mechanism allows a system administrator to select preferred host identifiers for each virtual machine or a group of virtual machines. The host identifiers mechanism in conjunction with an optimizer periodically adjusts the placement of virtual machines according to the preferred host identifiers and consistent with optimization policies. Further, the preferred host identifiers may include a time component so that the preferred host resources can be periodically updated and can dynamically change over time.

Placement of virtual machines on preferred physical hosts

Disclosed aspects manage virtual machine migration on a shared pool of configurable computing resources. A virtual machine is monitored in order to identify a set of migration data with respect to the virtual machine. A set of migration events is detected with respect to the virtual machine. Based on the set of migration events, the set of migration data is collected. In response to a triggering event, a determination is made whether to migrate the virtual machine from a current host based on the set of migration data. In accordance with the determination, a selection can be made whether to migrate the virtual machine from the current host.

Growing community of inventors

Austin, TX, United States of America

Taylor D Peoples