公开(公告)号：US20200264692A1

公开(公告)日：2020-08-20

申请号：US16642694

申请日：2017-09-28

申请人： Intel Corporation

发明人： Eugene GORBATOV ,  Alexander UAN-ZO-LI ,  Chee Lim NGE ,  James HERMERDING, II ,  Zhongsheng WANG

IPC分类号： G06F1/3296 ,  G06F1/28

摘要： Peak power setting circuitry is provided to set a peak power value for an integrated circuit device. A power supply interface is to receive a value to estimate a peak power capacity of a power supply serving the integrated circuit device and processing circuitry is provided to calculate an approximate peak power for the integrated circuit device. A peak power for the integrated circuit device is determined by increasing the approximate peak power depending on an amount by which the integrated circuit device power is reduced in response to assertion of a throttling signal.

公开(公告)号：US20220187893A1

公开(公告)日：2022-06-16

申请号：US17442374

申请日：2020-07-14

申请人： INTEL CORPORATION

发明人： Premanand SAKARDA ,  Efraim ROTEM ,  Eliezer WEISSMANN ,  Hisham ABU SALAH ,  Hadas BEJA ,  Russell FENGER ,  Deepak GANAPATHY ,  James HERMERDING, II ,  Ido KARAVANY ,  Nivedha KRISHNAKUMAR ,  Sudheer NAIR ,  Gilad OLSWANG ,  Moran PERI ,  Avishai WAGNER ,  Zhongsheng WANG ,  Noha YASSIN

IPC分类号： G06F1/324 ,  G06F1/3296

摘要： Described are mechanisms and methods for tracking user behavior profile over large time intervals and extracting observations for a user usage profile. The mechanisms and methods use machine learning (ML) algorithms embedded into a dynamic platform and thermal framework (DPTF) (e.g., Dynamic Tuning Technology) and predict device workloads using hardware (HW) counters. These mechanisms and methods may accordingly increase performance and user responsiveness by dynamically changing an Energy Performance Preference (EPP) based on a longer time workload analysis and workload prediction.