摘要:
Technologies to monitor and manage platform, device, processor and power characteristics throughout a system utilizing a remote entity such as controller node. By remotely monitoring and managing system operation and performance over time, future system performance requirements may be anticipated, allowing system parameters to be adjusted proactively in a more coordinated way. The controller node may monitor, control and predict traffic flows in the system and provide performance modification instructions to any of the computer nodes and a network switch to better optimize performance. The target systems collaborate with the controller node by respectively monitoring internal resources, such as resource availability and performance requirements to provide necessary resources for optimizing operating parameters of the system. The controller node may collect local system information for one or all of the computer nodes to dynamically steer traffic to a specific set of computers for processing to meet desired performance and power requirements.
摘要:
A method and apparatus for managing core affinity for network packet processing is provided. Low-power idle state of a plurality of processing units in a system including the plurality of processing units is monitored. Network packet processing is dynamically reassigned to processing units that are in a non-low power idle state to increase the low-power idle state residency for processing units that are in a low-power idle state resulting in reduced energy consumption.
摘要:
In an embodiment, an apparatus is provided that may include circuitry to generate, at least in part, and/or receive, at least in part, at least one request that at least one network node generate, at least in part, information. The information may be to permit selection, at least in part, of (1) at least one power consumption state of the at least one network node, and (2) at least one time period. The at least one time period may be to elapse, after receipt by at least one other network node of at least one packet, prior to requesting at least one change in the at least one power consumption state. The at least one packet may be to be transmitted to the at least one network node. Of course, many alternatives, modifications, and variations are possible without departing from this embodiment.
摘要:
Technologies for a distributed hardware queue manager include a compute device having a procesor. The processor includes two or more hardware queue managers as well as two or more processor cores. Each processor core can enqueue or dequeue data from the hardware queue manager. Each hardware queue manager can be configured to contain several queue data structures. In some embodiments, the queues are addressed by the processor cores using virtual queue addresses, which are translated into physical queue addresses for accessing the corresponding hardware queue manager. The virtual queues can be moved from one physical queue in one hardware queue manager to a different physical queue in a different physical queue manager without changing the virtual address of the virtual queue.
摘要:
In an embodiment, an apparatus is provided that may include circuitry to generate, at least in part, and/or receive, at least in part, at least one request that at least one network node generate, at least in part, information. The information may be to permit selection, at least in part, of (1) at least one power consumption state of the at least one network node, and (2) at least one time period. The at least one time period may be to elapse, after receipt by at least one other network node of at least one packet, prior to requesting at least one change in the at least one power consumption state. The at least one packet may be to be transmitted to the at least one network node. Of course, many alternatives, modifications, and variations are possible without departing from this embodiment.
摘要:
In an embodiment, an apparatus is provided that may include circuitry to generate, at least in part, and/or receive, at least in part, at least one request that at least one network node generate, at least in part, information. The information may be to permit selection, at least in part, of (1) at least one power consumption state of the at least one network node, and (2) at least one time period. The at least one time period may be to elapse, after receipt by at least one other network node of at least one packet, prior to requesting at least one change in the at least one power consumption state. The at least one packet may be to be transmitted to the at least one network node. Of course, many alternatives, modifications, and variations are possible without departing from this embodiment.
摘要:
Methods and systems to balance the load among a set of processing units, such as servers, in a manner that allows the servers periods of low power consumption. This allows energy efficient operation of the set of processing units. Moreover, the process is adaptable to variations in systemic response times, so that systemic response times may be improved when operational conditions so dictate.
摘要:
A power management unit (PMU) may determine an optimal power saving state using a break-even period of a power saving state and an expected idle duration based on a first policy. The PMU may determine the optimal power saving state using a first break even period and actual idle duration based on a second policy. The break-even period may equal a minimum time a computer system should remain in a power saving state to compensate for the power consumed by the system to enter and exit that power saving state. The expected idle time duration is determined as an average of idle duration and a recent sample of idle duration. The actual idle duration is the difference of a first and second time point that represents entry and exit points to and from the power saving state. The PMU may transition the system to the optimal power saving state.
摘要:
In an embodiment, an apparatus is provided that may include circuitry to generate, at least in part, and/or receive, at least in part, at least one request that at least one network node generate, at least in part, information. The information may be to permit selection, at least in part, of (1) at least one power consumption state of the at least one network node, and (2) at least one time period. The at least one time period may be to elapse, after receipt by at least one other network node of at least one packet, prior to requesting at least one change in the at least one power consumption state. The at least one packet may be to be transmitted to the at least one network node. Of course, many alternatives, modifications, and variations are possible without departing from this embodiment.
摘要:
Methods and systems to balance the load among a set of processing units, such as servers, in a manner that allows the servers periods of low power consumption. This allows energy efficient operation of the set of processing units. Moreover, the process is adaptable to variations in systemic response times, so that systemic response times may be improved when operational conditions so dictate.