Abstract: Embodiments maintain high availability of software application instances in a fault domain. Subordinate hosts are monitored by a master host. The subordinate hosts publish heartbeats via a network and datastores. Based at least in part on the published heartbeats, the master host determines the status of each subordinate host, distinguishing between subordinate hosts that are entirely inoperative and subordinate hosts that are operative but partitioned (e.g., unreachable via the network). The master host may restart software application instances, such as virtual machines, that are executed by inoperative subordinate hosts or that cease executing on partitioned subordinate hosts.

Abstract: In one embodiment, a method for placing virtual machines in a collection is provided. A plurality of equivalence sets of hosts is determined prior to placing virtual machines in the collection. The hosts in an equivalence set of hosts are considered similar. An equivalence set of hosts in the plurality of equivalence sets is selected to place the virtual machines in the collection. The method then places at least a portion of the virtual machines in the collection on one or more hosts in the selected equivalence set of hosts.

Abstract: Resources are provisioned in an automated manner for shared services in a resource-on-demand system. A model representing an observed state of resources in the resource-on-demand system allocated to the shared services and a model representing a desired state of the shared services are stored. At least one policy applicable to provisioning the resources for the shared services is determined. The policy and information from the models are applied to automatically provision the resources for satisfying the desired state of the shared services.

Abstract: An automated provisioning system provisions computing resources for shared services in a resource-on-demand system. A number of virtual machines needed for a virtual machine production pool for a period of time to satisfy the computing demand for the shared services is determined from a policy. Also, a number of unassigned virtual machines needed for a virtual machine buffer pool is determined from the policy. Servers and virtual machines are automatically provisioned for the virtual machine production pool and the virtual machine buffer pool based on the determined number of virtual machines needed for the virtual machine production pool, the determined number of unassigned virtual machines needed for the virtual machine buffer pool, and a changing computing demand of the shared services during the period of time.

Abstract: A computer system for maximizing system and individual job throughput includes a number of computer hardware processor cores that differ amongst themselves in at least in their respective resource requirements and processing capabilities. A monitor gathers performance metric information from each of the computer hardware processor cores that are specific to a particular run of application software then executing. Based on these metrics, a workload assignment mechanism assigns jobs to processor cores in order to maximize overall system throughput and the throughput of individual jobs.

Abstract: Resources are provisioned for shared services in a resource-on-demand system. A model representing an observed state of resources in the resource-on-demand system allocated to the shared services and a model representing a desired state of the shared services are stored. At least one policy applicable to provisioning the resources for the shared services is determined. The policy and information from the models are applied to automatically provision the resources for satisfying the desired state of the shared services.

Abstract: In a method of allocating workload among servers based upon historical data, a requested workload profile is received and is compared with a plurality of historical workload profiles. A historical workload profile that is within a predefined range of the requested workload profile and that corresponds to a substantially minimized resource usage for power and cooling is selected. In addition, workload among the servers is allocated according to the selected historical workload profile to thereby substantially minimize resource usage for power and cooling.

Abstract: According to one embodiment, a method comprises providing a machine-readable meta-model that defines the structure of how information is represented in at least one data model. The method further comprises using, by a data accessor, the meta-model for interpreting the at least one data model. According to another embodiment, a method comprises providing a machine-readable metric meta-model that defines a syntax for defining metric models, and defining a metric model in the syntax defined by the metric meta-model. The method further comprises associating the metric model with a monitoring source in a monitoring environment, wherein the metric model defines monitoring data available at the monitoring source, and interpreting, by a monitoring tool, the metric model based on the metric meta-model.

Abstract: According to one embodiment, a method comprises storing metric definitions for at least one monitored component in a machine-readable format to a data storage device. The method further comprises enabling access by at least one monitoring tool to the metric definitions via a metric introspection interface. The method further comprises the at least one monitoring tool autonomously comprehending the metric definitions for use in processing monitoring data collected for the at least one monitored component.

Abstract: According to one embodiment of the present invention, a method comprises providing a reporting network for communicating data among parts of a monitoring architecture as desired, wherein the reporting network is dynamically configurable programmatically. The method further comprises maintaining a machine-readable model of the monitoring architecture, and autonomously adapting configuration of the reporting network based on the machine-readable model.

Abstract: According to one embodiment of the present invention, a method comprises providing a machine-readable monitoring model that maintains configuration of a monitoring environment. An element of the monitoring environment reads the machine-readable monitoring model and adapts its operation to the configuration defined thereby.

Abstract: According to one embodiment, a method comprises providing a metric reporting configuration interface for enabling configuration of metrics included in monitoring data collected for at least one monitored component. The method further comprises supporting, by the metric reporting configuration interface, defining of configuration parameters of at least one metric to be reported in monitoring data collected for the at least one monitored component. The method further comprises collecting monitoring data for the at least one monitored component, and reporting the monitoring data in accordance with the defined configuration parameters.

Abstract: In a method of allocating workload among servers based upon historical data, a requested workload profile is received and is compared with a plurality of historical workload profiles. A historical workload profile that is within a predefined range of the requested workload profile and that corresponds to a substantially minimized resource usage for power and cooling is selected. In addition, workload among the servers is allocated according to the selected historical workload profile to thereby substantially minimize resource usage for power and cooling.

Abstract: A computer system for conserving operating power includes a number of computer hardware processor cores that differ amongst themselves in at least in their respective operating power requirements and processing capabilities. A monitor gathers performance metric information from each of the computer hardware processor cores that is specific to a particular run of application software then executing. A workload transfer mechanism transfers the executing application software to a second computer hardware processor core in a search for reduced operating power. A transfer delay mechanism is connected to delay a subsequent transfer of the executing application software if the system operating power may be conserved by such delay.

Abstract: A method for operating a data center with a robotic device. In the method, a condition is detected in a location of the data center. The robotic device, which includes a camera and a manipulator, is maneuvered to travel to the location of the data center. The location of the data center is imaged with the camera of the robotic device and an object is manipulated with the manipulator of the robotic device.

Abstract: According to at least one embodiment, a method comprises identifying at least one causal path that includes a node of a distributed computing environment that is of interest. The method further comprises analyzing the identified at least one causal path to determine at least one time interval when the node is active in such causal path, and correlating consumption of a resource by the node to the node's activity in the at least one causal path.

Abstract: Power system components are mapped using a controller and a data collection system. The controller is operable to transmit a first power system component ID for the first power system component to a second power system component, wherein the second power system component receives power from the first power system component. The data collection system is operable to receive from the second power system component the first power system component ID and a second power system component ID for the second power system component. The data collection system is further operable to associate the first power system component with the second power system component based on the received IDs.

Abstract: Load demands in a power system are managed by determining whether load demands on one or more power system components need to be varied. If the load demands need to be varied, new load demands to be placed on the power system components are determined. The load demands on the power system components are controlled such that the load demands are substantially equal to the new load demands.

Abstract: A plurality of computer systems is operable to execute an application. A power consumption is determined for a first computer system if the first computer system were to execute the application, and a power consumption is determined for at least one other computer system if the second computer system were to execute the application. One of the computer systems is selected to execute the application based on the determined power consumptions.

Abstract: A primary power supply and a secondary power supply are operable to supply power to at least one electrical device. A power demand of the electrical device is determined, and an efficient operating point for the primary power supply is determined. One or more of the primary power supply and the secondary power supply are used to supply power to the electrical device based on whether the primary power supply operating at an efficient operating point is operable to meet the power demand of the at least one electrical device.