System and Method to Measure Indirect Emissions Caused by Workload Hosted on a Distributed Cloud

    公开(公告)号:US20240362064A1

    公开(公告)日:2024-10-31

    申请号:US18140093

    申请日:2023-04-27

    IPC分类号: G06F9/48

    CPC分类号: G06F9/4893

    摘要: A system, method, and computer-readable medium for performing a data center management and monitoring operation. The data center management and monitoring operation includes: receiving data center asset data for a plurality of data center assets, the data center asset data comprising data associated with servicing a particular workload during a particular interval of time by a particular data center asset; monitoring electrical power consumed by a particular data center asset; and, performing an electrical power consumption analysis operation, the electrical power consumption analysis operation ascertaining an amount of electrical power consumed by the particular data center asset when servicing the particular workload during the particular interval of time. In certain embodiments, the power consumption analysis operation computes carbon emission generated by the particular workload.

    DEVICE TASK MANAGEMENT IN A DISTRIBUTED SYSTEM

    公开(公告)号:US20240330051A1

    公开(公告)日:2024-10-03

    申请号:US18473653

    申请日:2023-09-25

    IPC分类号: G06F9/48 H04L67/10

    CPC分类号: G06F9/4893 H04L67/10

    摘要: In aspects of device task management in a distributed system, the distributed system includes a task-source device, a compute device, and a task-target device. The compute device receives a task processing request to process a task from the task-source device. The compute device implements a task processing manager that obtains a power availability state for each device in the distributed system of the devices. The task processing manager can also obtain a thermal state for each device in the distributed system of the devices. The task processing manager selects, based on the power availability state and/or the thermal state for each device, the task-target device from the distributed system of the devices. The task processing manager then initiates to provide at least a portion of the task to the task-target device for processing by the task-target device.

    SYSTEMS AND METHODS FOR MANAGING POWER CONSUMPTION OF A DISTRIBUTED SYSTEM

    公开(公告)号:US20240220386A1

    公开(公告)日:2024-07-04

    申请号:US18609884

    申请日:2024-03-19

    摘要: Systems, apparatuses, methods, and computer program products are disclosed for managing operation of a distributed system that provides computer-implemented services. An example method includes collecting, via an out-of-band stream, sensor data from a plurality of temperature sensors positioned to infer temperatures of components of a distributed system. The example method further includes inferring, using the inferred temperatures, power consumption rates for performing respective actions with each of the components. The method further includes obtaining an action request and selecting, based on the inferred power consumption rates, a particular component of the components to perform an action to service the action request, the selection being made with a preference to reduce power consumption for performance of the action. The method further includes performing, with the particular component, the action to service the action request, the action resulting in updated computer-implemented services being provided by the distributed system.

    Memory device for scheduling maximum number of memory macros write operations at re-arranged time intervals

    公开(公告)号:US12026404B2

    公开(公告)日:2024-07-02

    申请号:US18313374

    申请日:2023-05-08

    IPC分类号: G06F3/06 G06F1/28 G06F9/48

    摘要: A memory device including a memory array with a plurality of memory macros, a power supplying circuit, and a controller is provided. The power supplying circuit is coupled to the memory array. The controller is coupled to the memory array. The power supplying circuit is configured to provide power to perform write operations to a number of the memory macros at the same time. The number of the memory macros for the write operations performed at the same time is not higher than a maximum number of the memory macros. The controller obtains the maximum number of the memory macros for the write operations performed at the same time by the power supplying circuit. The controller re-arranges a schedule for a sequence of the write operations of the memory macros to generate a re-arranged schedule. The maximum number is taken as a threshold value. In the re-arranged schedule, a number of part of the memory macros for the write operations performed at the same time is equal to or less then the threshold value.

    THERMAL CONTROLS BASED ON POWER ADDERS IN HETEROGENEOUS COMPUTING PLATFORMS

    公开(公告)号:US20240192757A1

    公开(公告)日:2024-06-13

    申请号:US18062757

    申请日:2022-12-07

    IPC分类号: G06F1/324 G06F1/20 G06F9/48

    摘要: Systems and methods for thermal controls based on power adders in heterogenous computing platforms are described. In an illustrative, non-limiting embodiment, an Information Handling System (IHS) may include a heterogeneous computing platform including a plurality of devices and a memory coupled to the heterogeneous computing platform, where the memory includes a plurality of sets of firmware instructions, and where at least one of the plurality of devices operates as an orchestrator configured to: in response to detection of a power adder coupled to the IHS, change a sustained power limit usable by a selected one of the plurality of devices.