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

Stream scheduling method and device based on stable matching, computer device and storage medium

A matching method and flow scheduling technology, applied in digital transmission systems, data exchange networks, electrical components, etc., can solve problems such as inability to ensure correctness, coarse-grained bandwidth allocation, and Fincher's inability to adapt to topology.

Pending Publication Date: 2019-06-11
GUANGZHOU UNIVERSITY
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, even if their assumptions are true for the fat-tree topology, they cannot be guaranteed to be true for other topologies
In other words, Fincher cannot be adapted to all types of topologies
Also, the downside of Fincher's flow-to-switch matching is coarse-grained bandwidth allocation (whose granularity is all bandwidth between two switches)

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
  • Stream scheduling method and device based on stable matching, computer device and storage medium
  • Stream scheduling method and device based on stable matching, computer device and storage medium
  • Stream scheduling method and device based on stable matching, computer device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] This embodiment provides a flow scheduling method based on stable matching, which divides the link bandwidth of each hop into some network slices, and constructs a certain number of paths by combining network slices according to the number of flows, and each path is only connected to one Streams are matched, and each stream matches exactly one path. Therefore, this embodiment transforms the flow scheduling into a classic one-to-one stable matching problem, such as the stable marriage problem.

[0057] Such as figure 1 As shown, the flow scheduling method based on stable matching in this embodiment is aimed at an elephant flow EFN of unknown size, which includes the following steps:

[0058] S101. Set a stream and a path.

[0059] In the flow scheduling method based on stable matching in this embodiment, two objects need to be set: flow and path.

[0060] Set streams: A set of streams is defined as F = {f 1 , f 2 ,...}, where |F| is the total number of flows. Will ...

Embodiment 2

[0087] Such as figure 2 As shown, this embodiment provides a stream scheduling device based on stable matching, which includes a setting module 201, a preference table building module 202, a stable matching model building module 203, and a stream scheduling module 204. The specific functions of each module are as follows:

[0088] The setting module 201 is used to set streams and paths;

[0089] The preference table construction module 202 is configured to construct a preference table for streams and paths;

[0090] The stable matching model construction module 203 is configured to construct a flow-path stable matching model according to the flow and path preference tables;

[0091] The flow scheduling module 204 is used to solve the optimal solution according to the flow-path stable matching model, and use the one-to-one stable matching method to find the best one-to-one stable matching between the flow and the path, so as to realize the optimal flow scheduling.

[0092] ...

Embodiment 3

[0094] This embodiment provides a computer device, which may be a server, such as image 3 As shown, it includes a processor 302 connected through a system bus 301, a memory and a network interface 303, the processor is used to provide computing and control capabilities, the memory includes a non-volatile storage medium 304 and an internal memory 305, the non-volatile The volatile storage medium 304 stores an operating system, computer programs, and databases. The internal memory 305 provides an environment for the operation of the operating system and computer programs in the nonvolatile storage medium. When the processor 302 executes the computer programs stored in the memory, the above-mentioned The flow scheduling method based on stable matching in embodiment 1 is as follows:

[0095] set stream and path;

[0096] Build preference tables for streams and paths;

[0097] According to the flow and path preference table, construct a flow-path stable matching model;

[0098]...

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 discloses a flow scheduling method and device based on stable matching, a computer device and a storage medium. The method comprises the following steps of setting a flow and a path; Constructing a preference table for the flow and the path; constructing the flow-path preference tables according to flow and path, constructing a flow-path stability matching model according to the flow-path preference tables; determining an optimal solution according to the flow-path stability matching model; and adopting a one-to-one stable matching method to find the optimal one-to-one stable matching between the flow and the path, so that the optimal scheduling of the flow is realized. According to the method, the defects of special topology oriented, coarse bandwidth distribution granularity and the like existing in an existing flow scheduling mechanism are overcome, the fine-grained bandwidth distribution capable of being applied to all types of topologies is achieved, and the simulation experiment results prove that the method has great advantages compared with previous work.

Description

technical field [0001] The invention relates to a flow scheduling method, device, computer equipment and storage medium based on stable matching, and belongs to the field of flow scheduling of a software-defined data center network. Background technique [0002] Currently, software-defined data-center networks (Software-Defined Data-Center Networks, SD-DCN) have been widely deployed, such as Google's B4. There are three types of flows in DCN, namely Mice Flow (MF for short), Elephant Flow with known sizes (EFY for short) and Elephant Flow with unknown size (Elephant Flow with unknown sizes). sizes, EFN). For example, some user-oriented data center applications (web search, social network, recommendation system, etc.) will generate MF; applications such as virtual machine migration and data backup will generate elephant flows of known size; while database access and HTTP-based large Applications such as chunk transfers generate elephant streams of unknown size. More import...

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): H04L12/801H04L12/721H04L12/24
Inventor 刘外喜沈湘平陈庆春唐冬卢锦杰刘晓初
Owner GUANGZHOU UNIVERSITY
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