摘要:
Various approaches for the deployment and coordination of inter-satellite communication pathways, defined for use with a satellite non-terrestrial network, are discussed. Among other examples, such inter-satellite communication pathways may be identified, reserved, allocated, and used for ultra-low-latency communication purposes.
摘要:
Methods for frequency scaling for per-core accelerator assignments and associated apparatus. A processor includes a CPU (central processing unit) having multiple cores that can be selectively configured to support frequency scaling and instruction extensions. Under this approach, some cores can be configured to support a selective set of AVX instructions (such as AVX3/5G-ISA instructions) and/or AMX instructions, while other cores are configured to not support these AVX/AMX instructions. In one aspect, the selective AVX/AMX instructions are implemented in one or more ISA extension units that are separate from the main processor core (or otherwise comprises a separate block of circuitry in a processor core) that can be selectively enabled or disabled. This enables cores having the separate unit(s) disabled to consume less power and/or operate at higher frequencies, while supporting the selective AVX/AMX instructions using other cores. These capabilities enhance performance and provides flexibility to handle a variety of applications requiring use of advanced AVX/AMX instructions to support accelerated workloads.
摘要:
A method according to one embodiment may include discovering, by software, at least one variable from at least one component populated on a shelf system. The method may also include performing, by the software, at least one shelf management function based on at least one variable. Of course, many alternatives, variations, and modifications are possible without departing from this embodiment.
摘要:
Described herein are a method and an apparatus for reducing power consumption of memories by monitoring the power states of the memories via an operating system. The method comprises reading counter values corresponding to power states of each memory of a plurality memories; computing a power state usage corresponding to the power states of each memory of the plurality, the computing based on the counter values; determining whether the power state usage exceeds a predetermined threshold usage; and adjusting current and future usage of each memory of the plurality in response to determining that the power state usage exceeds the predetermined threshold usage.
摘要:
A method according to one embodiment may include discovering, by software, at least one variable from at least one component populated on a shelf system. The method may also include performing, by the software, at least one shelf management function based on at least one variable. Of course, many alternatives, variations, and modifications are possible without departing from this embodiment.
摘要:
A hardware management module is enabled to perform hardware management for a modular platform system that includes a plurality of modular platform shelves coupled via one or more communication links in a network. Hardware management to include monitoring board interfaces resident on one or more backplanes within the plurality of modular platform shelves, detecting when a board is received and coupled to a board interface and performing one or more hardware management functions to include obtaining field replaceable unit information from the detected board.
摘要:
A method is to include implementing at least one statistical prediction model to predict memory power utilization and reduce power consumption for a computing platform. The implementation includes determining a configuration parameter for the computing platform, monitoring an operating parameter for the computing platform and predicting memory power utilization for the computing platform based on the determined configuration parameter and the monitored operating parameter. The method is to also include transitioning at least one memory module resident on the computing platform to one of a plurality of power states based at least in part on memory power utilization predicted via the implementation of the at least one statistical prediction model.
摘要:
A method for managing a system includes monitoring a plurality of applications running in the system for errors. A prediction is made as to whether errors detected would result in a failure. Fault recovery is initiated in response to a failure prediction. According to one aspect of the present invention, monitoring the plurality of applications includes reading error recorders associated with error occurrence. Other embodiments are described and claimed.
摘要:
A system is disclosed that includes a component, a fault table configured to receive fault information associated with the component, and a diagnosis processor configured to read the fault information from the fault table and initiate corrective action as a function of the fault information. A method for handling faults in the system is also disclosed.
摘要:
A method to reduce memory power consumption for a computing platform includes inspecting an operating parameter associated with a resource of the computing platform that is updated by the resource of the computing platform during runtime of the computing platform. Memory power utilization is then predicted for the computing platform during the runtime of the computing platform based at least in part on the operating parameter. A current power state of at least one memory module resident on the computing platform is transitioned to one of a plurality of power states based on the predicting of the memory power utilization.