Search

US-20260126848-A1 - Power and Thermal Management for Multiple Processing Cores

US20260126848A1US 20260126848 A1US20260126848 A1US 20260126848A1US-20260126848-A1

Abstract

A method and apparatus for power and thermal management for multiple processing cores are disclosed. Existing dynamic voltage and frequency scaling (DVFS) approaches typically perform localized management for individual cores, such as a central processing unit (CPU) or a graphics processing unit (GPU), resulting in suboptimal overall power consumption and thermal throttling. The disclosed energy manager addresses this by performing energy management across at least two computing cores executing a hybrid workload with a shared execution deadline. The energy manager acquires historical performance data to predict the execution time and total power consumption for a plurality of voltage and frequency combinations. An optimal combination is selected that minimizes the total predicted power consumption while meeting the deadline. The system further determines a thermal constraint, such as a thermal power budget, and restricts the available voltage and frequency combinations, ensuring continuous operation within safe thermal limits while improving power efficiency.

Inventors

  • Guibing Cai
  • Wei Wang
  • Sayanna Chandula

Assignees

  • GOOGLE LLC

Dates

Publication Date
20260507
Application Date
20251218

Claims (20)

  1. 1 . A method comprising: receiving a hybrid workload with an execution deadline; identifying, based on timeline data and power consumption data, a plurality of voltage and frequency combinations for a first computing core and a second computing core by which to complete the hybrid workload within the execution deadline; and adjusting, based on a selected one of the voltage and frequency combinations, an operating voltage and frequency of the first computing core and the second computing core, the adjustment effective to cause the first computing core and the second computing core to complete the hybrid workload within the execution deadline.
  2. 2 . The method of claim 1 , wherein the first computing core is a central processing unit (CPU) and the second computing core is a graphics processing unit (GPU).
  3. 3 . The method of claim 2 , wherein the timeline data and power consumption data further includes data associated with a third computing core, the third computing core comprising a tensor processing unit (TPU) or a neural processing unit (NPU).
  4. 4 . The method of claim 1 , wherein the identifying includes determining a predicted total completion time using the first computing core and the second computing core and comparing the predicted total completion time to the execution deadline.
  5. 5 . The method of claim 2 , wherein the identifying includes estimating a power cost difference for the CPU that is calculated based on: a latest measured power of a core cluster associated with the CPU; a change in a power efficiency due to the adjusting the operating voltage and frequency to the selected one of voltage and frequency combinations; and a ratio of total processing cycles for active hybrid workload tasks relative to total processing cycles for a target duration of the core cluster.
  6. 6 . The method of claim 5 , wherein the power efficiency is estimated based on: a previous operating efficiency of the CPU; and a new operating efficiency of the CPU.
  7. 7 . The method of claim 6 , wherein the identifying further includes estimating a total power consumption by summing the estimated power cost difference for the CPU and an estimated power cost difference for the GPU.
  8. 8 . The method of claim 1 , wherein the adjusted operating voltage and frequency results in the first computing core and the second computing core collectively consuming a lower power than at least one of the plurality of voltage and frequency combinations.
  9. 9 . An apparatus comprising: a first computing core; a second computing core; a memory configured to store a hybrid workload and historical timeline data; and an energy management module configured to: receive a hybrid workload with an execution deadline; identify, based on timeline data and power consumption data, a plurality of voltage and frequency combinations for a first computing core and a second computing core by which to complete the hybrid workload within the execution deadline; and adjust, based on a selected one of the voltage and frequency combinations, an operating voltage and frequency of the first computing core and the second computing core.
  10. 10 . The apparatus of claim 9 , wherein the first computing core is a central processing unit (CPU) and the second computing core is a graphics processing unit (GPU).
  11. 11 . The apparatus of claim 10 , wherein the energy management module includes an internal hybrid energy model configured to calculate a total predicted power consumption.
  12. 12 . The apparatus of claim 10 , further comprising a thermal management unit (TMU) coupled to the energy management module, the TMU configured to provide temperature data to the energy management module.
  13. 13 . The apparatus of claim 9 , wherein the energy management module is further configured to predict the completion time for the hybrid workload by determining the first computing core's execution time, the second computing core's execution time, and an overlap period between the first computing core's execution time and the second computing core's execution time.
  14. 14 . The apparatus of claim 9 , further comprising a third computing core, the third computing core comprising a tensor processing unit (TPU) or a neural processing unit (NPU).
  15. 15 . The apparatus of claim 9 , wherein the energy management module is configured to perform the adjustment by signaling a dynamic voltage and frequency scaling (DVFS) controller to adjust the operating voltage and frequency of the first computing core and the second computing core.
  16. 16 . A method comprising: receiving a hybrid workload with an execution deadline; determining a thermal constraint for the hybrid workload based on at least one of temperature data and a thermal power budget; identifying, based on timeline data and power consumption data, a plurality of voltage and frequency combinations for a first computing core and a second computing core by which to complete the hybrid workload within the execution deadline, where the plurality of voltage and frequency combinations is constrained by the thermal constraint; and adjusting, based on a selected one of the voltage and frequency combinations, an operating voltage and frequency of the first computing core and the second computing core.
  17. 17 . The method of claim 16 , wherein the determining a thermal constraint includes calculating a thermal headroom, wherein the thermal headroom is the difference between the execution deadline and a total frame processing time of the hybrid workload.
  18. 18 . The method of claim 17 , wherein the total frame processing time is calculated as the sum of the time spent by the first computing core and the second computing core minus an overlap period of the processing.
  19. 19 . The method of claim 17 , wherein the thermal constraint applies a frequency cap that limits the available execution deadline feasible voltage and frequency combinations.
  20. 20 . The method of claim 17 , wherein the thermal constraint is adjusted based on the thermal headroom being positive or negative, wherein when the headroom is positive, the constraint applies throttling, and when the headroom is negative, the constraint releases throttling.

Description

CROSS-REFERENCE TO RELATED APPLICATION This application claims the benefit of U.S. Provisional Patent Application Ser. No. 63/942,788 filed on Dec. 17, 2025, the disclosure of which is incorporated by reference herein in its entirety. SUMMARY A method and apparatus for power and thermal management for multiple processing cores are disclosed. Existing dynamic voltage and frequency scaling (DVFS) approaches typically perform localized energy management for individual cores, such as a central processing unit (CPU) or a graphics processing unit (GPU), resulting in suboptimal overall power consumption and thermal throttling. The disclosed energy manager addresses this by performing coordinated voltage and frequency management across at least two computing cores executing a hybrid workload with a shared execution deadline. The energy manager acquires historical performance data to predict the execution time and total power consumption for a plurality of voltage and frequency combinations. An optimal combination is selected that minimizes the total predicted power consumption while meeting the deadline. The system further determines a thermal constraint, such as a thermal power budget, and restricts the available voltage and frequency combinations, ensuring continuous operation within safe thermal limits while improving power efficiency. This document describes techniques and apparatuses, implemented on computing devices (e.g., mobile phones, tablets, and gaming consoles), for power and thermal management for multiple processing cores. In modern computing devices, particularly mobile platforms with multiple processing cores, tasks are often split across different types of cores, such as a CPU and a GPU. These hybrid workloads, like rendering a user interface (UI) frame, share a common performance deadline. Conventional power management systems often manage each core independently using dynamic voltage and frequency scaling (DVFS). This can lead to suboptimal power consumption for the system as a whole, as the coordination needed to meet the shared deadline most efficiently is lacking. For instance, one core might operate at a higher, less efficient frequency than necessary, increasing overall power draw and device temperature. This can trigger thermal throttling, which may degrade the user experience by causing missed deadlines and reduced frame rates. This system is designed to provide power efficiency and performance by coordinating the operation of heterogeneous computing cores (e.g., a CPU and a GPU) and integrating the resulting control decisions with thermal limits. In aspects, the present disclosure relates to a method for power and thermal management for multiple processing cores. The method includes receiving a hybrid workload with an execution deadline. The method further includes identifying, based on timeline data and power consumption data, a plurality of voltage and frequency combinations for a first computing core and a second computing core by which to complete the hybrid workload within the execution deadline. The method also includes adjusting, based on a selected one of the voltage and frequency combinations, an operating voltage and frequency of the first computing core and the second computing core, the adjustment effective to cause the first computing core and the second computing core to complete the hybrid workload within the execution deadline. In aspects, the present disclosure relates to a thermal-aware method. The method includes receiving a hybrid workload with an execution deadline. The method includes determining a thermal constraint for the hybrid workload based on at least one of temperature data and a thermal power budget. The method includes identifying, based on timeline data and power consumption data, a plurality of voltage and frequency combinations for a first computing core and a second computing core by which to complete the hybrid workload within the execution deadline, where the plurality of voltage and frequency combinations is constrained by the thermal constraint. The method includes adjusting, based on a selected one of the voltage and frequency combinations, an operating voltage and frequency of the first computing core and the second computing core based on a selected one of the voltage and frequency combinations. This document also describes aspects that may include one or more of the following features. In aspects, the first computing core may be a CPU, and the second computing core may be a GPU. In aspects, the timeline data may include an overlap period during simultaneous core execution. In further aspects, the timeline data and power consumption data may further include data associated with a third computing core, the third computing core comprising a tensor processing unit (TPU) or a neural processing unit (NPU). In aspects, the identifying may include determining a predicted total completion time using the first computing core and the second computing core