Abstract: Methods and apparatus to provide holistic global performance and power management are described. In an embodiment, logic (e.g., coupled to each compute node of a plurality of compute nodes) causes determination of a policy for power and performance management across the plurality of compute nodes. The policy is coordinated across the plurality of compute nodes to manage a job to one or more objective functions, where the job includes a plurality of tasks that are to run concurrently on the plurality of compute nodes. Other embodiments are also disclosed and claimed.

Abstract: Systems, apparatuses and methods may provide for obtaining, by a system level reallocator in a plurality of reallocators arranged in a hierarchical tree, resource budget information. Additionally, application performance information may be obtained by at least one of the plurality of reallocators. Moreover, a performance imbalance between a plurality of compute subtrees associate with the application performance information may be reduced by the at least one of the plurality of reallocators and based at least in part on the resource budget information and the application performance information.

Abstract: Various embodiments comprise prioritizing frequency allocations in thermally- or power-constrained computing devices. Computer elements may be assigned ‘weights’ based on their priorities. The computer elements with higher weights may receive higher frequency allocations to assure they receive priority in processing more quickly. The computer elements with lower weights may receive lower frequency allocations and suffer a slowdown in their processing. Elements with the same weight may be grouped together for the purpose of frequency allocation.

Abstract: An apparatus is provided which comprises: a component; a voltage generator to supply load current to the component; first one or more circuitries to predict that the load current is to increase from a first time; and second one or more circuitries to, in anticipation of the increase in the load current from the first time, cause the component to execute first instructions during a time period that occurs prior to the first time.

Abstract: An apparatus system is provided which comprises: a first component and a second component; a first circuitry to assign the first component to a first group of components, and to assign the second component to a second group of components; and a second circuitry to assign a first maximum frequency limit to the first group of components, and to assign a second maximum frequency limit to the second group of components, wherein the first component and the second component are to respectively operate in accordance with the first maximum frequency limit and the second maximum frequency limit.

Abstract: An apparatus is provided which comprises: a controller to allocate, to a component, a resource budget selected from a plurality of quantization levels; and a circuitry to adaptively update the plurality of quantization levels.

Abstract: Technologies for dynamic bandwidth management of interconnect fabric include a compute device configured to calculate a predicted fabric bandwidth demand which is expected to be used by the interconnect fabric in a next epoch and subsequent to a present epoch. The compute device is additionally configured to determine whether any global links and/or local links of the interconnect fabric can be disabled during the next epoch as a function of the calculated predicted fabric bandwidth demand and a number of redundant paths associated with the links of the interconnect fabric. The compute device is further configured to disable one or more of the global links and/or the local links that can be disabled, the one or more local links of the plurality of local links that can be disabled. Other embodiments are described herein.

Abstract: Apparatus and methods may provide for subscribing a thread to a resource monitor through a machine specific register and subscribing the thread to a class of service through the machine specific register. The resource monitor or the class of service for the thread may be changed without interrupting the thread. The power allocated to the processor core may be changed based on the selected class of service for the thread.

Abstract: Methods and apparatus to provide holistic global performance and power management are described. In an embodiment, logic (e.g., coupled to each compute node of a plurality of compute nodes) causes determination of a policy for power and performance management across the plurality of compute nodes. The policy is coordinated across the plurality of compute nodes to manage a job to one or more objective functions, where the job includes a plurality of tasks that are to run concurrently on the plurality of compute nodes. Other embodiments are also disclosed and claimed.

Abstract: Methods and apparatus to provide holistic global performance and power management are described. In an embodiment, logic (e.g., coupled to each compute node of a plurality of compute nodes) causes determination of a policy for power and performance management across the plurality of compute nodes. The policy is coordinated across the plurality of compute nodes to manage a job to one or more objective functions, where the job includes a plurality of tasks that are to run concurrently on the plurality of compute nodes. Other embodiments are also disclosed and claimed.

Abstract: Apparatus, systems, and methods provide an interface between a plurality of hardware resources of a node and a power manager. The interface is configured to define one or more resource groups to expose to the power manager for power measurement and control, assign the plurality of hardware resources to the one or more resource groups, and provide a power allowance to each resource group.

Abstract: Apparatus, systems, and methods provide dynamic spatial power steering among a plurality of power domains in the computer system on a per phase basis of a particular application. Dynamic spatial power steering may include, for example, determining a plurality of phases corresponding to an application comprising tasks for execution on a processing node. determining a spatial power distribution between a plurality of power domains on the processing node based on a performance metric for each phase, monitoring the application to detect a current phase, and applying the spatial power distribution correspond to the current phase to the plurality of power domains.

Abstract: Apparatus and methods may provide for characterizing a plurality of similar components of a distributed computing system based on a maximum safe operation level associated with each component and storing characterization data in a database and allocating non-uniform power to each similar component based at least in part on the characterization data in the database to substantially equalize performance of the components.

Abstract: Apparatus, systems, and methods provide dynamic power steering that includes determining a sequence of phases of an application in a node. The sequence corresponds to a time interval associated with an energy budget. For each phase, the dynamic power steering includes determining a power scaling comprising a measured response to an increase or decrease in power distributed to a plurality of power domains in the node, and based on the power scaling for each phase, determining a temporal power distribution between the phases in the sequence to satisfy the energy budget.

Abstract: Apparatus, systems, and methods provide dynamic spatial power steering among a plurality of power domains in the computer system on a per phase basis of a particular application. Dynamic spatial power steering may include, for example, determining a plurality of phases corresponding to an application comprising tasks for execution on a processing node. determining a spatial power distribution between a plurality of power domains on the processing node based on a performance metric for each phase, monitoring the application to detect a current phase, and applying the spatial power distribution correspond to the current phase to the plurality of power domains.

Abstract: Apparatus and methods may provide for subscribing a thread to a resource monitor through a machine specific register and subscribing the thread to a class of service through the machine specific register. The resource monitor or the class of service for the thread may be changed without interrupting the thread. The power allocated to the processor core may be changed based on the selected class of service for the thread.

Abstract: Systems, apparatuses and methods may provide for obtaining, by a system level reallocator in a plurality of reallocators arranged in a hierarchical tree, resource budget information. Additionally, application performance information may be obtained by at least one of the plurality of reallocators. Moreover, a performance imbalance between a plurality of compute subtrees associate with the application performance information may be reduced by the at least one of the plurality of reallocators and based at least in part on the resource budget information and the application performance information.

Abstract: Systems, apparatuses and methods may provide for a plurality of node-level agents, wherein each node-level agent aggregates network statistics information from a plurality of probes associated with a communications interface. Additionally, one or more job-level agents may be communicatively coupled to the plurality of node-level agents, wherein each job-level agent aggregates network statistics information from two or more of the node-level agents. Moreover, a system-level agent may be communicatively coupled to the job-level agent(s). The system-level agent may generate a power model based on aggregated network statistics information from the job-level agent(s) and propagate the power model to the node-level agents via the job-level agent(s).

Abstract: Apparatus, systems, and methods provide an interface between a plurality of hardware resources of a node and a power manager. The interface is configured to define one or more resource groups to expose to the power manager for power measurement and control, assign the plurality of hardware resources to the one or more resource groups, and provide a power allowance to each resource group.

Abstract: Apparatus and methods may provide for characterizing a plurality of similar components of a distributed computing system based on a maximum safe operation level associated with each component and storing characterization data in a database and allocating non-uniform power to each similar component based at least in part on the characterization data in the database to substantially equalize performance of the components.