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

Multi-data-center energy-saving routing method and system based on reinforcement learning

A multi-data center and reinforcement learning technology, applied in digital transmission systems, transmission systems, data exchange networks, etc.

Active Publication Date: 2020-12-04
SHANDONG UNIV
View PDF17 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In order to solve the shortcomings of the existing technology, the present disclosure provides a multi-data center energy-saving routing method and system based on reinforcement learning, which solves the weight selection problem of energy consumption and bandwidth cost, obtains a high-quality topology solution set, and then realizes the best energy-saving effect

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
  • Multi-data-center energy-saving routing method and system based on reinforcement learning
  • Multi-data-center energy-saving routing method and system based on reinforcement learning
  • Multi-data-center energy-saving routing method and system based on reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] Such as figure 1 As shown, Embodiment 1 of the present disclosure provides a multi-data center energy-saving routing method based on reinforcement learning, including the following steps:

[0051] In the multi-data center network topology based on software-defined network architecture, for elastic transmission requirements, under the premise of ensuring load balancing, use reinforcement learning to find a topology subset that meets both low energy consumption and low bandwidth costs;

[0052]Using the nonlinear characteristics of the Chebyshev scalar function, optimize the optional action set in reinforcement learning, obtain the scalarized ε greedy strategy and replace the original action selection strategy, so as to solve the weight of energy consumption and bandwidth cost problem, obtain a high-quality solution set;

[0053] After the topology subset is finally selected, unselected link devices can be turned off, so as to meet transmission requirements at low cost a...

Embodiment 2

[0090] Embodiment 2 of the present disclosure provides a multi-data center energy-saving routing system based on reinforcement learning, including:

[0091] The node set acquisition module is configured to: for any transmission requirement, initialize the current transmission location as the starting point of the route, and obtain an optional node set for the next transmission;

[0092] The node selection module is configured to: select the next node from the set of optional nodes using a scalarized ε greedy strategy;

[0093] The node confirmation module is configured to: obtain the transition probability from the current node to the next node according to the link congestion factor, judge whether to select the node as the next node according to the comparison between the transition probability and the preset threshold, and if not, continue the node selection, If so, the selected node is used as the new routing starting point to continue routing node selection until the trans...

Embodiment 3

[0096] Embodiment 3 of the present disclosure provides a medium on which a program is stored, and when the program is executed by a processor, the steps in the reinforcement learning-based multi-data center energy-saving routing method described in Embodiment 1 of the present disclosure are implemented.

[0097] The detailed steps are the same as the reinforcement learning-based multi-data center energy-saving routing method provided in Embodiment 1, and will not be repeated here.

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 multi-data-center energy-saving routing method and system based on reinforcement learning. The method comprises the steps of: initializing a current transmission position as arouting starting point for any transmission demand to obtain an optional node set for next transmission; selecting a next node from the optional node set by adopting a scalarized epsilon greedy strategy; acquiring a transition probability from a current node to a next node according to a link congestion factor, judging whether to select the node as the next node or not according to comparison between the transition probability and a preset threshold value, if not, further performing node selection, and if so, further performing routing node selection by taking the selected node as a new routing starting point till a transmission end point; according to the invention, the problem of weight selection of energy consumption and bandwidth cost is solved, a high-quality solution set is obtained, and a comprehensive optimal topology subset is selected from the solution set to realize an optimal energy-saving effect.

Description

technical field [0001] The present disclosure relates to the field of data center network transmission, in particular to an energy-saving routing method and system for multi-data centers based on reinforcement learning. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] In recent years, more and more attention has been paid to green energy saving. As a network infrastructure, while the data center continues to develop, the problem of energy consumption is also becoming more and more serious. The study found that the energy consumption of network transmission is mainly concentrated in routing equipment such as routers and switches, which account for more than half of the total energy consumption of the equipment, while the energy consumption of links such as optical fibers accounts for less than 1%. [0004] In terms of data center energy ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): H04L12/709H04L12/721H04L12/729H04L12/751H04L45/243H04L45/02H04L45/125
CPCH04L45/08H04L45/245H04L45/125H04L45/124
Inventor 王华燕嘉鑫伊善文
Owner SHANDONG 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