CN104468212B - A kind of cloud computation data center network intelligence linkage collocation method and system - Google Patents
A kind of cloud computation data center network intelligence linkage collocation method and system Download PDFInfo
- Publication number
- CN104468212B CN104468212B CN201410728270.4A CN201410728270A CN104468212B CN 104468212 B CN104468212 B CN 104468212B CN 201410728270 A CN201410728270 A CN 201410728270A CN 104468212 B CN104468212 B CN 104468212B
- Authority
- CN
- China
- Prior art keywords
- server
- virtual machine
- link
- data center
- busy percentage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of cloud computation data center network intelligence linkage collocation method and system, it is related to technical field of computer network management, this method includes：Extract the resource operation information and resource allocation information in the cloud computation data center network；According to the resource operation information and the resource allocation information, analysis obtains the status information and the status information of link of server in the cloud computation data center network；It is handled as follows according to the status information of the server and the status information of the link：When the cpu busy percentage exception of the server or the abnormal utilization rate of the link, it is adjusted by the position for migrating virtual machine in the server.The present invention effectively solves current cloud computing Visualized data centre network strategy and configures dumb, the problem of complexity is high and can not perceive cloud computing platform resource utilization.
Description
Technical field
The present invention relates to technical field of computer network management, be one kind according to cloud platform resource utilisation information, Intelligent joint
The method for completing the adjustment of cloud computation data center network configuration dynamicly.
Background technology
Cloud computing (Cloud Computing), as the latest developments of current information technology, just changes subtle
The production and life of human society.It is built for resource and the unified pipe of task by technologies such as Distributed Calculation, virtualizations
The Resource Control Layer of scheduling is managed, scattered ICT resources are put together to form resource pool, dynamic on-demand is distributed to apply and used.
In recent years, with the fast development of cloud computing technology, on the one hand, search, social networks, multimedia audio-video, scientific algorithm, number
According to traditional " big data " business such as excavations, warp-wise cloud platform is migrated；On the other hand, more many enterprise start deployment and
Implement cloud computing strategy.
The automation and intellectuality of network management are the inexorable trends of cloud computation data center development.With cloud computing technology
Fast development, calculate, storage, the virtualization of network have been widely used, but these resources particularly Internet resources
Automation and intelligent management are but still within the starting stage.Current cloud computing Visualized data centre network strategy, which also exists, matches somebody with somebody
Put dumb, the problem of complexity is high and can not perceive cloud computing platform resource utilization.
Data center network (Data Center Network, DCN), sends out as the underlying infrastructure of cloud computing platform
Wave irreplaceable important function.It is used to connect large-scale physics and virtual server, and is upper strata cloud computing application
High-throughput, low latency, multipath and manageable communication service are provided.In recent years, with national government support energetically and
The progressively implementation of enterprise's cloud strategy, cloud computation data center is largely built.Current cloud computation data center network shows
New feature following aspects：(1) data center network popularization and number of devices increase sharply.(2) data center is virtual
Change degree is improved constantly.(3) data center network discharge model changes.Above-mentioned new feature gives current cloud computation data center network
Monitoring management and O＆M optimization bring unprecedented challenge, be mainly reflected in the aspect of following protrusion：
Network security and isolation strategy are dumb, in current virtualization cloud computation data center network, network security
With isolation strategy typically by two layers of VLAN, three straton nets, the network equipment level accesses control list and firewall rule etc.
Traditional approach is implemented.These safety and isolation strategy from traditional IP govern in current virilization cloud computing platform and provided
The flexibility of source dynamic configuration and real-time migration.
Network configuration management complexity is high, although virtual machine can be by current cloud computing management platform (such as
OpenStack, EC2 etc.) dynamic on-demand distribution, but the key-course path of corresponding physical network and virtual network and
Network strategy then need one by one equipment go configuration, this process is not only dry as dust, and easily malfunctions.
The resource utilization of cloud computing platform can not be perceived, the utilization of resources feelings of cloud computing platform are considered due to lacking
Condition, the independent management to multi-tenant virtual network easily causes the unbalanced of whole cloud platform resource utilization, and when cloud is flat
When platform resource utilization changes, it is impossible to realize and the tuning of virtual switch network strategy is configured.
Patent of invention " a kind of data center's method of automatic configuration and its equipment " is applied to two layers of internet of data center,
It is provided with being configured with the VPN for supporting two layers of forwarding on address distribution apparatus and address resolution equipment, each core network device,
Virtual two layers of connection is set up between each core network device, this method includes：Address distribution apparatus receives data center's edge device
The logon message sent after the IP address configuration of complete capital equipment, wherein carrying the MAC Address and IP of the edge device
Address；Address distribution apparatus caches the address information of the data center's edge device carried in the logon message, and passes through core
Virtual two layers of connection of heart net equipment room, the MAC Address of the edge device is sent out in the form of address cluster is VPN-MAC route
Cloth.The present invention with can causing the IP of the edge device of data center in data center's generation migration, operator's change of access etc.
When location changes, the carry out data center configuration of dynamic self-adapting, without manual intervention.The patent can automatically configure number
According to the MAC Address and IP address of center network apparatus, but on the one hand the method for the patent does not provide in configuration data and has a guilty conscience
Intend the method for network strategy；On the other hand it reasonably can not adjust network using the resource utilisation information counted in cloud platform
Configuration, collocation method is dumb.
Patent of invention " virtual network management system and method in cloud computation data center " (number of patent application
201210130269.2), it is related to network architecture and network communication protocol technical field.The system includes：Physical server,
It is connected with data center core switching network, possesses at least one interchanger for supporting OpenFlow related protocols, and physics thereon
Server is connected with controller；Controller, builds virtual network, safeguards the configuration of virtual network, and virtual network with it is virtual wide
The mapping relations between domain and local broadcast domain are broadcast, the interchanger on one or more physical server is configured and controlled
System.The system and method for the present invention can efficiently reduce data center's core switching network broadcast and L2 address table and forward table
Pressure, it is extensive beneficial to realizing；Meanwhile, it is capable to which the flow of different user is effectively isolated, strengthen security；And
The virtual network configuration of data center is set rapidly flexibly to be changed with the DYNAMIC DISTRIBUTION of virtual resource.
This method has following two defects：1. network configuration management complexity is high, not only dry as dust, Er Qierong
It is error-prone；2. can not consider the overall resource utilization of cloud platform, the independent management to multi-tenant virtual network is easy
The unbalanced of whole cloud platform resource utilization is caused, and when cloud platform resource utilization changes, it is impossible to realize to void
Intend the tuning configuration of switch network strategy.
The present invention considers the overall running situation of cloud platform from global angle, and the utilization of resources in cloud is perceived in time
Situation of change, and intelligent linkage to cloud platform overall network configuration be adjusted so that cloud platform overall operation efficiency keep
It is optimal.
Invent " a kind of data center management system towards cloud computing ", be related to a kind of data center towards cloud computing and manage
Reason system, it by:Universal document system API module, monitoring resource module, resource alarm module, tactful configuration module, rule are drawn
Hold up, scheduling of resource module, performing module and resource virtualizing api interface module are constituted；Universal document system API module and resource
Scheduler module is electrically connected；Monitoring resource module is electrically connected with resource alarm module, scheduling of resource module respectively with monitoring resource mould
Block, resource alarm module, regulation engine, performing module electrical connection；Tactful configuration module is electrically connected with regulation engine；Performing module
Electrically connected with resource virtualizing api interface module.The data center management system towards cloud computing is to data center calculation
Rational management is carried out with store tasks, data center's intelligent scheduling is realized, improves the service quality of data center, improve reliability
Reduce operation maintenance cost.
This method fails to be managed cloud computation data center with reference to SDN (software defined network) technology so that net
Network configuration strategy underaction, it is impossible to configured for contents such as network equipment flow tables.
Intelligent linkage of the present invention research based on software defined network distributes technology rationally, wherein in cloud computing data
The tackling key problem of plan of having a guilty conscience network automatically configuring management method helps to solve the problems, such as the operation management after network virtualization；And for
The tackling key problem for carrying out intelligent optimum operation based on cloud platform resource utilisation information, which is then advantageously implemented, really regard network as one kind money
Source control gets up.Finally, be conducive to improving the operation management level of virtualization cloud computation data center, lifting cloud platform resource profit
With rate, the favourable service quality for ensureing cloud business.
The content of the invention
In order to solve the above problems, the present invention proposes the cloud computation data center network intelligence based on software defined network
Link configuring technical.The resource utilisation information of whole cloud platform is obtained by current cloud computing management platform first, in this base
By software defined network technology rational management and distributing the Internet resources in cloud platform rationally on plinth, multi-tenant demand is being met
While improve the service efficiency of whole cloud platform system resource.
The present invention provides a kind of cloud computation data center network intelligence linkage collocation method, comprises the following steps：
Step 1：Extract the resource operation information and resource allocation information in the cloud computation data center network；
Step 2：According to the resource operation information and the resource allocation information, analysis obtains the cloud computation data center network
The status information of middle server and the status information of link；
Step 3：It is handled as follows according to the status information of the server and the status information of the link：
It is empty in the server by migrating when the cpu busy percentage exception of the server or the abnormal utilization rate of the link
The position of plan machine is adjusted.
Described cloud computation data center network intelligence linkage collocation method, the step 3 also includes：
When the server cpu busy percentage is higher than 80%, the maximum virtual machine of cpu busy percentage in the server is searched, by this
Virtual machine (vm) migration is less than 40% and topology location and the nearest new demand servicing device of the server to cpu busy percentage, until the server
Cpu busy percentage be less than 70% untill；
When the server cpu busy percentage is less than 20%, the virtual machine (vm) migration in the server is less than to cpu busy percentage
40% and the nearest new demand servicing device of topology location and the server, and by the server dormancy；
When the new demand servicing device cpu busy percentage is higher than 60%, by the startup of server under resting state, and by the new demand servicing
Virtual machine (vm) migration in device is to the server, untill the server cpu busy percentage is higher than 20%；
Described cloud computation data center network intelligence linkage collocation method, the step 3 also includes；
When the utilization rate of the link is higher than 70%, inquiry is by the large data stream of the link, and by the large data stream
Starting point virtual machine (vm) migration to the terminal virtual machine of the large data stream where server under frame, and make the clothes under the frame
The cpu busy percentage of business device is between 20% to 80%.
Described cloud computation data center network intelligence linkage collocation method, the step 3 also includes：When virtual machine will be disposed
In the cloud computation data center network, by the deploying virtual machine in the minimum server of cpu busy percentage；And analyze the service
The hot-spot link of device connection, makes the data flow of the virtual machine bypass the hot-spot link.
The present invention also proposes a kind of cloud computation data center network intelligence linkage configuration system, including：
Utilization of resources module, believes for extracting the resource operation information in the cloud computation data center network with resource distribution
Breath；
Running state analysis module, for according to the resource operation information and the resource allocation information, analysis to obtain the cloud
Calculate the status information and the status information of link of server in data center network；
Resource intelligent linkage adjusting module, is carried out for the status information according to the server and the status information of the link
Following processing：
It is empty in the server by migrating when the cpu busy percentage exception of the server or the abnormal utilization rate of the link
The position of plan machine is adjusted.
Described cloud computation data center network intelligence linkage configuration system, resource intelligent linkage adjusting module is also wrapped
Include：
When the server cpu busy percentage is higher than 80%, the maximum virtual machine of cpu busy percentage in the server is searched, by this
Virtual machine (vm) migration is less than 40% and topology location and the nearest new demand servicing device of the server to cpu busy percentage, until the server
Cpu busy percentage be less than 70% untill；
When the server cpu busy percentage is less than 20%, the virtual machine (vm) migration in the server is less than to cpu busy percentage
40% and the nearest new demand servicing device of topology location and the server, and by the server dormancy；
When the new demand servicing device cpu busy percentage is higher than 60%, by the startup of server under resting state, and by the new demand servicing
Virtual machine (vm) migration in device is to the server, untill the server cpu busy percentage is higher than 20%；
Described cloud computation data center network intelligence linkage configuration system, resource intelligent linkage adjusting module is also wrapped
Include；
When the utilization rate of the link is higher than 70%, inquiry is by the large data stream of the link, and by the large data stream
Starting point virtual machine (vm) migration to the terminal virtual machine of the large data stream where server under frame, and make the clothes under the frame
The cpu busy percentage of business device is between 20% to 80%.
Described cloud computation data center network intelligence linkage configuration system, resource intelligent linkage adjusting module is also wrapped
Include：When virtual machine will be deployed in the cloud computation data center network, by the deploying virtual machine in the minimum service of cpu busy percentage
In device；And the hot-spot link of server connection is analyzed, the data flow of the virtual machine is bypassed the hot-spot link.
Technique effect：The technique effect of cloud platform Network resource allocation intelligent linkage configuration seeks to the money according to cloud platform
Source is using the resource requirement information of information and business in cloud, and rational scheduling and the Internet resources configured in cloud platform are more to meet
Tenant's demand and the service efficiency for improving resource, the present invention propose that the intelligent linkage based on software defined network distributes skill rationally
Art, can carry out automatic configuration management for cloud computation data center virtual network, help to solve after network virtualization
Operation management problem；The priority scheduling of resource of intelligence can be carried out based on cloud platform resource utilisation information, is advantageously implemented net
Network really gets up as a kind of resource management, and overall having the technical effect that can be improved entirely while multi-tenant demand is met
The service efficiency of cloud platform system resource, is conducive to improving the operation management level of virtualization cloud computation data center, lifts cloud
Platform resource utilization rate, the favourable service quality for ensureing cloud business.
Brief description of the drawings
Fig. 1 is intelligent linkage application effect figure；
Fig. 2 is intelligent linkage embodiment figure；
Fig. 3 is utilization of resources illustraton of model.
Embodiment
The present invention needs to obtain the resource of cloud platform by the existing interface of cloud computing platform (such as OpenStack, EC2)
Using information, such as physical server CPU/ memory usages, data center network traffic matrix and other cloud platform performances
Information etc.；Physically or a virtually interchanger in cloud computation data center network needs to support software defined network, such as existing
OpenFlow agreements；Need independently to dispose a controller node in cloud computing platform, the node can either be with cloud computing pipe
Platform communicate, can also with cloud platform support SDN (software defined network technology) equipment communication, cloud computing can be used as
The plug-in unit of management platform, can also individualism, for carrying out distributing rationally for intelligent linkage.
(1) cloud computing platform resource utilisation information model
In order to be intelligently managed to the Internet resources in cloud platform, it is necessary to grasp the global operation ginseng of cloud platform in real time
Number and resource service condition, comprehensive ground of planning as a whole carry out centralized management to Internet resources.Cloud computing platform resource utilisation information model,
Using resource running situation in the current cloud of accurate description as target, define it is a set of can completely, accurately reflect resource status in cloud
Index system, abundant full and accurate resource utilisation information is provided for the most optimum distribution of resources decision-making on upper strata.Cloud computing platform resource profit
Mainly include two parts with information model, Part I is operational factor, includes the operational factor and physics money of virtual resource
These data are excavated, concluded, arranged by the operational factor in source, form the things such as network bandwidth utilization factor, network traffics size
Reason and virtual network resource performance indications, and the physics such as cpu busy percentage, memory usage and virtual computing resource performance refer to
Mark；Part II is configuration information, mainly for network configuration information, wherein both having included channel-group (passage group), void
Intend the virtual network configurations such as network port ACL, also including the physical network port configuration such as VLAN divisions, it is often more important that needs pair
Network configuration based on SDN technologies is acquired, such as key-course path, stream definition.
The foundation of cloud computing platform resource utilisation information model is mutually tied using independent excavate with cloud platform management data-reusing
The technology of conjunction, to the data existed in cloud platform, such as cpu busy percentage, is directly multiplexed, to SDN configuration, void
Intending network configuration etc. needs the independent data excavated, using call related management API or based on SSH call instruction interfaces by the way of
Obtain.
Technique effect：Cloud computing platform resource utilisation information model is capable of the resource utilisation information of highly effective gathering cloud platform,
Filtering, which is extracted, in the magnanimity informations such as slave unit operational factor, network performance index, flow information matrix can react that to portray cloud flat
The key message of platform current operating situation, and processing and sorting distributes the model with directive significance into one to upper resource
System, is the follow-up decision information enriched based on cloud platform real time resources using the intelligent linkage configuration provides of information.
(2) cloud platform Network resource allocation intelligent linkage is configured
The configuration of cloud platform Network resource allocation intelligent linkage mainly solves how to utilize information and each industry according to global resource
It is engaged in adjusting the demand information of resource the distribution of cloud platform Internet resources, on the premise of business demand is met, adjusts Internet resources
Configuration so that network performance is optimized.Means are configured first with traditional traffic shaping, bandwidth allocation etc. to distribute for each tenant
The resource needed, recycles the configuration means based on SDN such as traffic engineering to realize the optimization to the whole network performance.Meeting business need
On the premise of asking, optimize link utilization, alleviate network hotspot load pressure, congestion is avoided as far as possible, reach lifting resource profit
With efficiency, optimize the purpose of network performance.
Problem is abstracted into a mathematical modeling based on Multidimensional Knapsack Problems first, that is, needs to put M kind Internet resources
Enter in N number of business knapsack, meet the demand of all business knapsacks, the problem is solved using optimal method so that M kinds network is provided
The utilization rate in source is optimized.Wherein, optimization problem is shown below：
s.t.pi≤bi(i=1,2...n)
Wherein piThe Internet resources occupied by each business, n is the total number of business in cloud, biTo ensure that business is normal
The resource requirement of work.So that Internet resources distribution total amount it is minimum, but must be fulfilled for it for the resource of each traffic assignments
Demand.
Then the prediction algorithm based on high confidence level is needed, performance boost and adjustment that accurate estimation adjustment algorithm is brought
The resource consumption that process is caused in itself, when income consumes the algorithm triggers value than being higher than our predefineds, starts network money
Internet resources are redistributed by source intelligent linkage Regulation mechanism.In order to optimize the receipts for performing Internet resources Regulation mechanism
Benefit consumption ratio, using the performance boost situation after the method prediction adjustment of quantitative analysis, i.e. precise quantification analysis network configuration is adjusted
The influence that can be produced after whole to network, such as core link utilization rate, average bandwidth service condition are compared with current state
Compared with being then based on historical data and predict this resource consumption that can produce of adjustment.When ratio between two reaches one based on experience
During trigger value, start Internet resources intelligent linkage adjustment algorithm, the Internet resources of the whole network are reassigned.
It is below the specific workflow of the present invention, as shown in Figure 3：
Utilization of resources module, utilization of resources module is deployed in cloud platform as a single node, is responsible for extracting cloud
Information and resource allocation information during resource operation in platform, and pass to running state analysis module after being modeled and enter
Row analyzing and processing, utilization of resources module is divided into operational factor and configuration parameter two large divisions, and its specific composition form is one
XML file, form is as follows：
<runtime>
<virtual>
<name>vm name</name>
<vcpu>vcpu utilization</vcpu>
<vtraffic>virtual traffic</vtraffic>
</virtual>
<physical>
<name>name</name>
<cpu>cpu utilization</cpu>
<traffic>traffic</traffic>
</physical>
</runtime>
<configuration>
<virtual conf>
<target>vm name</target>
<vcpuconf>virtual cpu configuration</vcpuconf>
<vport>vport configuration</vport>
</virtual conf>
<physical conf>
<cpuconf>cpu configuration</cpuconf>
<networkconf>network configuration</networkconf>
</physical conf>
<SDN conf>
<SDNconf>SDN configuration</SDNconf>
</SDN conf>
</configuration>
Running state analysis module, running state analysis module is mainly responsible for the XML file analyze data by being collected into
The status information of central server and the status information of link, the analysis of server state information mainly include cpu busy percentage
Analysis, the related content that can directly extract XML file is obtained；The analysis of Link State needs to combine in utilization of resources module
Server traffic information and SDN information acquisitions, the size of data flow is obtained by server traffic information, is obtained by SDN information
To the trend of data flow, after running state analysis module completes to analyze, result is separately recorded in following three tables, chain
Road information table：The initial address of link is recorded, end address, link bandwidth, link utilization uses the server set of the link
Close；Server info table：Record the mac addresses of server, IP address, logical number, cpu busy percentage, cpu frequency；Virtual letter
Cease table：Record the UUID (numbering) of virtual machine, place server, virtual cpu quantity, CPU Expenditure Levels, virtual uninterrupted
Deng.
Resource intelligent linkage adjusting module, when running state analysis module finds that following two exceptions occurs in data center
When, it can be handled automatically：
Cpu busy percentage is abnormal：Certain server cpu busy percentage is higher than 80%：Find CPU consumptions maximum in the server
Virtual machine, is migrated to it, is less than 40% and topology location nearest server for cpu busy percentage move target, until
Untill the cpu busy percentage of the server is less than 70%；Certain server cpu busy percentage is less than 20%：Will be all in the server
Virtual machine (vm) migration is gone out, and is less than 40% and topology location nearest server for cpu busy percentage move target, is then taken this
Business device dormancy, reaches the purpose of energy-conservation；Server average CPU utilization is higher than 60%：By the startup of server under resting state,
And the virtual machine (vm) migration in the server by adjacent thereto and cpu busy percentage higher than 60% comes, until its cpu busy percentage is high
Untill 20%.
Link utilization is abnormal, and the utilization rate of certain link is higher than 70%：The large data stream by the link is found, and
By the server under frame where its starting point virtual machine (vm) migration to its terminal virtual machine, and ensure the CPU profits of selected server
With rate between 20% to 80%.
Resource intelligent linkage adjusting module, when there is new virtual machine to remove to be deployed in cloud computation data center network, money
Source intelligent linkage adjusting module is responsible for it and selects suitable server, and plans flow path by SDN technologies for it, specifically
Step is as follows：
Virtual machine tenant is analyzed, is its selection deployment frame (rack)；Find under the frame CPU resource utilization most
Low physical server, by the deploying virtual machine in the physical server；The hot-spot link of physical server connection is analyzed,
And SDN configurations are carried out to it, its flow is bypassed hot-spot link
Workflow when being accessed below for new tenant, as shown in Figure 2：
When new tenant accesses, running state analysis module analyzes its resource requirement situation first, then performs resource intelligence
Can be linked adjusting module, intelligent to distribute Internet resources for it, and complete configuration optimization, the network overall situation in cloud is kept optimal.
Timing of the invention obtains related operational factor from cloud computation data center network, and is obtained by utilization of resources module
Relevant parameter is taken, then transfers to running state analysis module to be analyzed, and Internet resources Intelligent joint is performed according to analysis result
Dynamic adjustment algorithm carries out tuning to the whole network performance.
Claims (6)
- The collocation method 1. a kind of cloud computation data center network intelligence links, it is characterised in that comprise the following steps：Step 1：Extract the resource operation information and resource allocation information in the cloud computation data center network；Step 2：According to the resource operation information and the resource allocation information, analysis is obtained in the cloud computation data center network and taken The status information of device of being engaged in and the status information of link；Step 3：It is handled as follows according to the status information of the server and the status information of the link：When the cpu busy percentage exception of the server or the abnormal utilization rate of the link, by migrating virtual machine in the server Position be adjusted；Wherein, the step 3 also includes：When the server cpu busy percentage is higher than 80%, the maximum virtual machine of cpu busy percentage in the server is searched, this is virtual Machine moves to cpu busy percentage and is less than 40% and topology location and the nearest new demand servicing device of the server, until the CPU of the server Untill utilization rate is less than 70%；When the server cpu busy percentage be less than 20%, by the virtual machine (vm) migration in the server to cpu busy percentage be less than 40% and The nearest new demand servicing device of topology location and the server, and by the server dormancy；When the new demand servicing device cpu busy percentage is higher than 60%, by the startup of server under resting state, and by the new demand servicing device Virtual machine (vm) migration to the server, untill the server cpu busy percentage is higher than 20%.
- The collocation method 2. cloud computation data center network intelligence as claimed in claim 1 links, it is characterised in that the step 3 Also include；When the utilization rate of the link is higher than 70%, inquiry passes through the large data stream of the link, and rising the large data stream Server under frame where point virtual machine (vm) migration to the terminal virtual machine of the large data stream, and make the server under the frame Cpu busy percentage between 20% to 80%.
- The collocation method 3. cloud computation data center network intelligence as claimed in claim 1 links, it is characterised in that the step 3 Also include：It is when virtual machine will be deployed in the cloud computation data center network, the deploying virtual machine is minimum in cpu busy percentage In server；And the hot-spot link of server connection is analyzed, SDN configurations are carried out, the data flow of the virtual machine is bypassed the heat Point link.
- The configuration system 4. a kind of cloud computation data center network intelligence links, it is characterised in that including：Utilization of resources module, for extracting resource operation information and resource allocation information in the cloud computation data center network；Running state analysis module, for according to the resource operation information and the resource allocation information, analysis to obtain the cloud computing The status information of server and the status information of link in data center network；Resource intelligent linkage adjusting module, is carried out as follows for the status information according to the server and the status information of the link Processing：When the cpu busy percentage exception of the server or the abnormal utilization rate of the link, by migrating virtual machine in the server Position be adjusted.Wherein, resource intelligent linkage adjusting module also includes：When the server cpu busy percentage is higher than 80%, the maximum virtual machine of cpu busy percentage in the server is searched, this is virtual Machine moves to cpu busy percentage and is less than 40% and topology location and the nearest new demand servicing device of the server, until the CPU of the server Untill utilization rate is less than 70%；When the server cpu busy percentage be less than 20%, by the virtual machine (vm) migration in the server to cpu busy percentage be less than 40% and The nearest new demand servicing device of topology location and the server, and by the server dormancy；When the new demand servicing device cpu busy percentage is higher than 60%, by the startup of server under resting state, and by the new demand servicing device Virtual machine (vm) migration to the server, untill the server cpu busy percentage is higher than 20%.
- The configuration system 5. cloud computation data center network intelligence as claimed in claim 4 links, it is characterised in that the resource intelligence The adjusting module that can link also includes；When the utilization rate of the link is higher than 70%, inquiry passes through the large data stream of the link, and rising the large data stream Server under frame where point virtual machine (vm) migration to the terminal virtual machine of the large data stream, and make the server under the frame Cpu busy percentage between 20% to 80%.
- The configuration system 6. cloud computation data center network intelligence as claimed in claim 4 links, it is characterised in that the resource intelligence The adjusting module that can link also includes：When virtual machine will be deployed in the cloud computation data center network, the deploying virtual machine is existed In the minimum server of cpu busy percentage；And the hot-spot link of server connection is analyzed, SDN configurations are carried out, make the virtual machine Data flow bypasses the hot-spot link.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410728270.4A CN104468212B (en) | 2014-12-03 | 2014-12-03 | A kind of cloud computation data center network intelligence linkage collocation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410728270.4A CN104468212B (en) | 2014-12-03 | 2014-12-03 | A kind of cloud computation data center network intelligence linkage collocation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104468212A CN104468212A (en) | 2015-03-25 |
| CN104468212B true CN104468212B (en) | 2017-08-08 |
Family
ID=52913622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410728270.4A Expired - Fee Related CN104468212B (en) | 2014-12-03 | 2014-12-03 | A kind of cloud computation data center network intelligence linkage collocation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104468212B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10341188B2 (en) * | 2015-01-27 | 2019-07-02 | Huawei Technologies Co., Ltd. | Network virtualization for network infrastructure |
| CN106301843A (en) * | 2015-05-28 | 2017-01-04 | 亿阳信通股份有限公司 | A kind of cloud platform safeguards system and method |
| CN105208099B (en) * | 2015-08-24 | 2018-12-11 | 浪潮(北京)电子信息产业有限公司 | A kind of interior architectural framework using the economize on electricity of SDN technical intelligence of Cloud Server |
| CN108337179B (en) * | 2017-01-19 | 2021-02-05 | 华为技术有限公司 | Link flow control method and device |
| US20180316626A1 (en) * | 2017-04-28 | 2018-11-01 | Futurewei Technologies, Inc. | Guided Optimistic Resource Scheduling |
| CN110149341B (en) * | 2019-05-29 | 2020-06-16 | 燕山大学 | Cloud system user access control method based on sleep mode |
| CN110515693A (en) * | 2019-07-26 | 2019-11-29 | 浪潮电子信息产业股份有限公司 | A kind of method and system that the virtual machine based on rack perception is extending transversely |
| CN111124692B (en) * | 2020-01-02 | 2023-05-12 | 神州数码融信软件有限公司 | Service request processing system |
| CN111753169B (en) * | 2020-06-29 | 2021-10-19 | 金电联行（北京）信息技术有限公司 | Data acquisition system based on internet |
| CN115314390B (en) * | 2022-06-23 | 2023-05-16 | 清华大学 | Cloud computing network measurement planning system and method supporting multiple modes |
| CN117008674B (en) * | 2023-10-07 | 2023-12-12 | 四川川西数据产业有限公司 | Intelligent monitoring and adjusting system for energy consumption of data center |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102710432A (en) * | 2012-04-27 | 2012-10-03 | 北京云杉世纪网络科技有限公司 | System and method for managing virtual network in cloud computation data center |
| CN102710509A (en) * | 2012-05-18 | 2012-10-03 | 杭州华三通信技术有限公司 | Automatic data center configuration method and method |
| CN103607459A (en) * | 2013-11-21 | 2014-02-26 | 东北大学 | Dynamic resource monitoring and scheduling method of cloud computing platform IaaS layer |
| CN104092565A (en) * | 2014-06-24 | 2014-10-08 | 复旦大学 | Multi-tenant policy-driven type software-defined networking method for cloud data center |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140229608A1 (en) * | 2013-02-14 | 2014-08-14 | Alcatel-Lucent Canada Inc. | Parsimonious monitoring of service latency characteristics |
-
2014
- 2014-12-03 CN CN201410728270.4A patent/CN104468212B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102710432A (en) * | 2012-04-27 | 2012-10-03 | 北京云杉世纪网络科技有限公司 | System and method for managing virtual network in cloud computation data center |
| CN102710509A (en) * | 2012-05-18 | 2012-10-03 | 杭州华三通信技术有限公司 | Automatic data center configuration method and method |
| CN103607459A (en) * | 2013-11-21 | 2014-02-26 | 东北大学 | Dynamic resource monitoring and scheduling method of cloud computing platform IaaS layer |
| CN104092565A (en) * | 2014-06-24 | 2014-10-08 | 复旦大学 | Multi-tenant policy-driven type software-defined networking method for cloud data center |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104468212A (en) | 2015-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104468212B (en) | A kind of cloud computation data center network intelligence linkage collocation method and system | |
| Zhang et al. | Cloud computing: state-of-the-art and research challenges | |
| CN105205231B (en) | A kind of power distribution network Digital Simulation System based on DCOM | |
| CN102457439B (en) | Virtual switching system and method of cloud computing system | |
| CN110662231A (en) | Network slice resource adjusting method and system for 5G environment | |
| Dai et al. | Enabling network innovation in data center networks with software defined networking: A survey | |
| Xu et al. | Enhancing kubernetes automated scheduling with deep learning and reinforcement techniques for large-scale cloud computing optimization | |
| Duan et al. | A load balancing and multi-tenancy oriented data center virtualization framework | |
| US20150156131A1 (en) | Method and system of geographic migration of workloads between private and public clouds | |
| CN106656867A (en) | Dynamic SDN (Software Defined Network) configuration method based on application awareness of virtual network | |
| CN105635283A (en) | Organization and management and using method and system for cloud manufacturing service | |
| CN105610715B (en) | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN | |
| CN103475722A (en) | Implement system for business collaboration platform | |
| CN111064649A (en) | Method and device for realizing binding of layered ports, control equipment and storage medium | |
| CN107547317A (en) | Virtualize control method, device and the communication system of BAS Broadband Access Server | |
| CN106899478A (en) | The method that power test business realizes resource resilient expansion by cloud platform | |
| CN104112049A (en) | P2P (peer-to-peer) architecture based cross-data-center MapReduce task scheduling system and P2P architecture based cross-data-center MapReduce task scheduling method | |
| Chen et al. | An sdn-based fabric for flexible data-center networks | |
| CN104539558A (en) | Capacity-expansible IP telephone exchange blade mechanism frame and automatic capacity expansion method | |
| CN109213566A (en) | Virtual machine migration method, device and equipment | |
| CN102571440B (en) | A kind of network management operation method and system | |
| CN104298539B (en) | Scheduling virtual machine and dispatching method again based on network aware | |
| CN106961440B (en) | Cloud platform based on the operation monitoring management of enterprise-level resource | |
| Chang et al. | Architecture design of datacenter for cloud english education platform | |
| CN115915404A (en) | Network slice deployment system and method based on NFV-MANO |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170808 Termination date: 20201203 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |