Adaptive packet scattering method based on path classification in data center network

A data center network, self-adaptive technology, applied in data exchange networks, digital transmission systems, electrical components, etc., can solve problems such as uneven traffic distribution, out-of-order data packets, inability to sense path congestion and rerouting in a timely manner, etc. To achieve the effect of improving the experience and reducing the completion time

Active Publication Date: 2018-12-18
CENT SOUTH UNIV +1
View PDF6 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The use of ECMP has the following problems: First, due to the use of hash functions for random route selection, the distribution of traffic between paths is uneven, and multiple long flows hashed to the same path are likely to cause "hot spots"
When using a large threshold, it can ensure that few packets are out of order, but it cannot sense path congestion and rerouting in time
When using a small threshold, there will be a packet out-of-order problem similar to the random packet scattering strategy

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
  • Adaptive packet scattering method based on path classification in data center network
  • Adaptive packet scattering method based on path classification in data center network
  • Adaptive packet scattering method based on path classification in data center network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The present invention will be further described below in conjunction with the accompanying drawings.

[0033] see figure 1 , the flow process of the present invention is as follows:

[0034] First, the number of outgoing ports n is initialized and set to the number of outgoing ports of the current switch; the path classification set C 1 to C n Set to empty; the current path classification set C cur Set to all paths; path detection period δ is set to 3 times the round-trip delay; queue difference threshold d is set to 3; path category number j is set to 1; the cache queue length corresponding to n outbound ports is k i Both are set to 0 to form a set K of outbound buffer queue lengths; the path p corresponding to n outbound ports i Both are set to the number of the path. Determine whether the detection period δ of the path detection timer is overtime, that is, whether the time spent in detecting the path exceeds the path detection period δ, if so, consider that the cu...

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 an adaptive packet scattering method based on path classification in a data center network. The exchanger classifies the exit paths corresponding to the exit ports according tothe length of the exit port buffer queue, and adopts random packet scattering technology to transmit the data packets of the same stream in the same class of exit paths. Compared with the method of transmitting data packets in the same stream in all exit paths by the random packet scattering technology, the invention can reduce the number of out-of-order packets in the data center network and thecompletion time of the stream, thereby improving the performance of the application and optimizing the user experience.

Description

technical field [0001] The invention relates to an adaptive packet scattering method based on path classification in a data center network. Background technique [0002] In recent years, with the continuous development of big data (Big Data) and cloud computing (Cloud Computing), the data center (Data Center) has gradually developed into the cornerstone of modern basic computing facilities. At present, a large number of Internet companies, such as Microsoft, Google, Amazon, and Alibaba, deploy online application systems in their data centers, and use the unique large-scale computing power and big data storage capabilities of data centers to provide users with various Internet service. The importance of the Data Center Network (DCN) as a link connecting large-scale servers in the data center for distributed computing is self-evident. [0003] The data center traffic has gradually developed from the traditional "north-south traffic" mode to the modern "east-west traffic" mod...

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): H04L12/803H04L12/835H04L12/861H04L12/935H04L47/30H04L49/111
CPCH04L47/125H04L47/30H04L49/30H04L49/90
Inventor 黄家玮李淑平蔡明远蔡荣吕文俊李琪
Owner CENT SOUTH UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap