Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Heterogeneous multi-core processor dynamic voltage regulation method based on reinforcement learning

A dynamic voltage regulation, multi-core processor technology, applied in control/regulation systems, regulation of electrical variables, instruments, etc., can solve the problem of inability to adapt to the power consumption and performance requirements of the processor platform, and achieve the effect of protecting technical details

Active Publication Date: 2021-07-23
SUN YAT SEN UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method cannot adapt to different types of processor platforms and meet different power consumption and performance requirements.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Heterogeneous multi-core processor dynamic voltage regulation method based on reinforcement learning
  • Heterogeneous multi-core processor dynamic voltage regulation method based on reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] This embodiment provides a method for dynamic voltage regulation of heterogeneous multi-core processors based on reinforcement learning, such as figure 1 , including the following steps:

[0060] S1: At each time step, obtain the state data of the multi-core processor, and obtain the state value after processing;

[0061] S2: Set the dynamic voltage adjustment command and get the action set;

[0062] S3: setting an agent, the agent selects a dynamic voltage adjustment command from the action set according to the state value at the current moment;

[0063] S4: The multi-core processor performs dynamic voltage adjustment according to the selected dynamic voltage adjustment instruction;

[0064] S5: Obtain the state value of the multi-core processor after dynamic voltage adjustment, and calculate the reward;

[0065] S6: Update the agent according to the reward, and select the dynamic voltage adjustment instruction for the next time step;

[0066] S7: Evaluate whether ...

Embodiment 2

[0089] A dynamic voltage regulation system for heterogeneous multi-core processors based on reinforcement learning, such as figure 2 , the system is based on the reinforcement learning-based dynamic voltage regulation method for heterogeneous multi-core processors described in Embodiment 1, including: a state value acquisition module, the state value acquisition module is used to acquire multi-core processing The status data of the device, and the status value is obtained after processing;

[0090] An action set acquisition module, the action set acquisition module is used to set a dynamic voltage adjustment command to obtain an action set;

[0091] An intelligent body setting module, the intelligent body setting module is used to set the intelligent body, and the intelligent body selects a dynamic voltage adjustment command from the action set according to the state value at the current moment;

[0092] A first adjustment module, the first adjustment module is configured to...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a heterogeneous multi-core processor dynamic voltage regulation method based on reinforcement learning, and the method comprises the following steps: S1, obtaining the state data of a multi-core processor at each time step, and processing the state data to obtain a state value; S2, setting a dynamic voltage regulation instruction to obtain an action set; S3, setting an intelligent agent, and selecting a dynamic voltage regulation instruction from the action set by the intelligent agent according to the state value at the current moment; S4, enabling the multi-core processor to perform dynamic voltage regulation according to the selected dynamic voltage regulation instruction; S5, obtaining a state value of the multi-core processor after dynamic voltage regulation, and calculating an award; S6, updating the intelligent agent according to the reward, and selecting a dynamic voltage regulation instruction of the next time step length; and S7, evaluating whether the dynamic voltage regulation achieves an expected effect or not, if so, ending the dynamic voltage regulation, and if not, returning to the step S3. The method can adapt to different types of processor platforms and meet different power consumption and performance requirements.

Description

technical field [0001] The invention relates to the field of multi-core processor voltage regulation, and more specifically, to a method and system for dynamic voltage regulation of heterogeneous multi-core processors based on reinforcement learning. Background technique [0002] Most of the existing dynamic voltage adjustment technologies perform global dynamic voltage frequency adjustment for all cores, or perform independent dynamic voltage frequency adjustment for each individual core. The hardware implementation of the global dynamic voltage frequency adjustment is simple, but the efficiency is low; the independent dynamic voltage frequency adjustment can reduce the energy consumption of the system well, but requires complex hardware support. Using multiple voltage domains can balance hardware overhead and energy efficiency. Heterogeneous multi-core processors usually put the same cores together to form a core cluster (cluster), share certain system resources (such as ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G05F1/625
CPCG05F1/625Y02D10/00
Inventor 陈弟虎粟涛杨旭明
Owner SUN YAT SEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com