Cloudcomputing network flow prediction method and device based on time sequence statistical segmentation

A cloud computing network and traffic forecasting technology, which is applied in the direction of data exchange network, transmission system, digital transmission system, etc., can solve problems such as network performance impact, adjacent traffic interference, and application performance stability, so as to improve accuracy, Optimize resource adjustment strategies, avoid scheduling and adjustment effects

Pending Publication Date: 2019-06-18
上海仪电(集团)有限公司中央研究院
View PDF1 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Due to the variety of services and applications carried in the cloud computing environment, and different applications have different requirements for network resources, in particular, in today's cloud computing environment where multiple applications coexist, a prominent problem is "adjacent traffic interference" (Noisy Neigbour) problem, that is, while sharing resources, due to insufficient isolation and scheduling of resources, some applications will interfere with other adjacent and related applications when they generate a large amount of resource requirements or generate a large amount of network traffic, affecting Performance stability for other applications
[0003] In order to solve the impact of adjacent traffic interference on other users on network resources, the current countermeasures mainly include virtual machine scheduling, network scheduling, resource isolation, etc., but these have a certain time overhead, that is, they cannot respond to network changes. real-time response, and may also have a certain impact on network performance itself. For example, virtual machine migration itself requires a certain amount of network bandwidth, so frequent adjustments cannot be made

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
  • Cloudcomputing network flow prediction method and device based on time sequence statistical segmentation
  • Cloudcomputing network flow prediction method and device based on time sequence statistical segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0027] Such as figure 1 As shown, a cloud computing network traffic prediction device based on time series statistics segmentation, including:

[0028] The traffic statistics module 1 is deployed in each server node of the cloud computing, and counts the change of the sending and receiving traffic of each virtual machine in the server node over time, and obtains the accurate time series of the network bandwidth usage of each virtual machine over time;

[0029] Sequence segmentation module 2. For traffic-time series, a sliding window mechanism is used to segment the unsegmented sequence in the window, so that the difference in the statistical characteristics of resource occupancy of each segment is not less than the set threshold, and then the window slides to the current Time point, continue to segment the unsegmented sequence;

[0030] Sequence statistics module 3, based on the learned time segmentation results, performs segment-based network bandwidth occupancy feature stat...

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 relates to a cloud computing network flow prediction method based on time sequence statistics segmentation, which comprises the following steps: carrying out time-phased resource occupation statistics on a virtual machine in a cloud computing network, establishing a Markov state model according to a statistical result, and carrying out network flow prediction in a next time period according to the model. Compared with the prior art, for non-periodic network behaviors, the method carries out time sequence data analysis and prediction on a network boundary bandwidth monitoring result, flow characteristics and time sequence trends of different services in the network are identified, and the possibility of network adjacent flow interference hidden troubles in a future period of time is judged, so that a resource adjustment strategy is optimized, and too frequent scheduling and adjustment are avoided.

Description

technical field [0001] The invention relates to a network traffic prediction method, in particular to a cloud computing network traffic prediction method and device based on time series statistics and segmentation. Background technique [0002] Due to the variety of services and applications carried in the cloud computing environment, and different applications have different requirements for network resources, in particular, in today's cloud computing environment where multiple applications coexist, a prominent problem is "adjacent traffic interference" (Noisy Neigbour) problem, that is, while sharing resources, due to insufficient isolation and scheduling of resources, some applications will interfere with other adjacent and related applications when they generate a large amount of resource requirements or generate a large amount of network traffic, affecting Performance stability for other applications. [0003] In order to solve the impact of adjacent traffic interferen...

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): H04L12/24H04L12/26
Inventor 张鹏飞
Owner 上海仪电(集团)有限公司中央研究院
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