A P2P Network Dynamic Load Balancing Method Based on Prediction and Region Division

A technology of P2P network and dynamic load, which is applied in the direction of data exchange network, digital transmission system, electrical components, etc., can solve the problem of no balance, and achieve the effect of reducing the ratio, reducing the ratio, and obviously gaining

Active Publication Date: 2021-05-11
HARBIN INST OF TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] For the existing load balancing algorithms according to the node load area, most of them adopt the strategy of combining actual nodes and virtual nodes, without taking into account the possible occurrence of the actual situation. In actual operation, there are seldom multiple virtual nodes running The case on the same real node

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
  • A P2P Network Dynamic Load Balancing Method Based on Prediction and Region Division
  • A P2P Network Dynamic Load Balancing Method Based on Prediction and Region Division
  • A P2P Network Dynamic Load Balancing Method Based on Prediction and Region Division

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0049] A P2P network dynamic load balancing method based on prediction and area division, such as figure 1 shown, including the following steps:

[0050] Step a, obtain node resource utilization;

[0051] Step b, judging whether the resource utilization of the node exceeds the threshold of the high-load area, if yes, execute step e, if not, execute step c;

[0052] Step c, predicting the next access volume according to the current file access situation of the node;

[0053] Step d, determine whether the file becomes a hot file, if so, execute step e, if not, execute step a;

[0054] Step e, broadcast to the surrounding nodes to obtain the load information of the surrounding nodes

[0055] Step f, judging whether it is in a high-load area, if so, execute the load-balancing module in the high-load area, if not, execute the load-balancing module in the low-load area;

[0056] Step g, the server creates a copy of the hotspot file, and proceeds to step a.

[0057] Such as figur...

specific Embodiment approach 2

[0078] When combining the triple exponential smoothing algorithm with the structured P2P network to predict hot files, each node needs to perform its own file heat calculation. The node periodically records the requested access status of the files in the current interval, and after recording, it is handed over to the algorithm for analysis. The current request result is above the threshold When , calculate the predicted value x of the hotspot file t+T , if the predicted value x t+T If it is greater than the hotspot file threshold Ψ (the threshold of the high-load area is different from the threshold of the low-load area, which is collectively denoted by Ψ), this node requests the load information of the surrounding nodes.

[0079] On the basis of the specific implementation mode 1, when the utilization rate of the node resource exceeds the threshold value of the high-load area, a local load balancing information collection method is adopted, including the following steps: ...

specific Embodiment approach 3

[0085] When a node enters a high-load state, that is, when the node utilization rate is higher than the threshold of the high-load area, the high-load node broadcasts K-hop routing information to the surrounding network, requests the load status of the surrounding nodes, and calculates whether the node is in the high-load area. When the surrounding nodes receive the load request information, the node detects the node utilization rate of its own node. If the node is in a low load state, that is, when the node utilization rate is less than the threshold of the high load area, the node returns the basic information of the node to the state requester, and returns A timestamp is used to calculate the link delay, and at the same time, the routing hop count K is reduced by one, and the request is broadcast to its surrounding nodes; if the node is in a high-load state, it only forwards the message after processing the routing hop count, and does not reply to the message.

[0086] On th...

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

A P2P network dynamic load balancing method based on prediction and area division belongs to the field of network file transmission; including obtaining node resource utilization; judging whether the node resource utilization exceeds the high load area threshold, if yes, execute step e, if not, execute step c; Predict the next access volume according to the current file access situation of the node; determine whether the file becomes a hot file, if so, execute step e, if not, execute step a; broadcast to the surrounding nodes to obtain the load information of the surrounding nodes; determine whether it is in a high Load area, if so, execute the high-load area load balancing module, if not, execute the low-load area load balancing module; the server creates a copy of the hotspot file, and proceeds to step a; the present invention can well realize the load balancing of the actual system network.

Description

technical field [0001] The invention belongs to the field of network file transmission, in particular to a P2P network dynamic load balancing method based on prediction and area division. Background technique [0002] Today, the rapid development of the Internet has become an indispensable part of life, and the amount of file distribution data through the P2P network is also increasing. Large-scale file distribution has a greater impact on the overall load balance of the P2P network. , so, for P2P networks, load balancing is a problem that needs to be solved. [0003] Although the load balancing strategy of the existing structured P2P network also involves hotspot prediction, it is not very clear about the selection of load migration nodes, and the performance of nodes is not comprehensively considered, and most of them pass the model network test performance. [0004] For the existing load balancing algorithms according to the node load area, most of them adopt the strate...

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 Patents(China)
IPC IPC(8): H04L29/08H04L12/24H04L12/26H04L12/733H04L12/803H04L45/122
CPCH04L41/145H04L41/147H04L43/16H04L45/20H04L47/125H04L67/104H04L67/1001
Inventor 叶麟余翔湛张宏莉刘飞扬赵俊达刘立坤顾旭张元禛葛蒙蒙
Owner HARBIN INST OF TECH
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