CN109039933A - A kind of cluster network optimization method, device, equipment and medium - Google Patents
A kind of cluster network optimization method, device, equipment and medium Download PDFInfo
- Publication number
- CN109039933A CN109039933A CN201810921913.5A CN201810921913A CN109039933A CN 109039933 A CN109039933 A CN 109039933A CN 201810921913 A CN201810921913 A CN 201810921913A CN 109039933 A CN109039933 A CN 109039933A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- destination server
- resource
- cluster network
- idling
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/29—Flow control; Congestion control using a combination of thresholds
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of cluster network optimization method, device, equipment and medium, the step of this method includes: the idling-resource information for obtaining destination server in cluster network；Wherein, the idle bandwidth of each network link of destination server is included at least in idling-resource information；When the value of the arbitrary target resource items in idling-resource information preset threshold corresponding less than target resource item, judge whether run virtual machine in destination server；If it is, according to the first logic of propositions in destination server selection target virtual machine, and according to the second logic of propositions select destination server；Using idle bandwidth as the constraint of volume of transmitted data, target virtual machine is transmitted to destination server.This method avoids cluster network during migration server virtual machine and generates network congestion, and then ensure that the availability and service quality of cluster network.In addition, the present invention also provides a kind of cluster network optimization device, equipment and medium, beneficial effect are same as above.
Description
Technical field
The present invention relates to internet areas, more particularly to a kind of cluster network optimization method, device, equipment and medium.
Background technique
Currently, as the cloud computing technology under big data background is fast-developing, the processing of data is gradually partial to
It is carried out by way of network cluster, the server in network cluster is no longer individual computing module, but needs network collection
Server in group cooperates, and therefore, when being monitored management to the server in network in this scenario, generally requires root
According to the overall condition of server, regulated and controled on a large scale.
By taking the data center that current enterprise generally uses as an example, data center is the control core of cloud computing framework, its energy
The service condition of the resources such as each server CPU, memory, storage, network under the statistics current cluster network of enough macroscopic views, and can be right
Resource load in each server carries out unified allocation of resources, and realization optimizes cluster network, to reach each in cluster network
The relatively average purpose of the operating pressure of server.
Since under current use pattern, a physical server would generally run multiple virtual machines, and with virtual machine
Support is provided for upper layer application for unit, therefore it is relatively small to loading to load the virtual machine (vm) migration in biggish server
It is to load to carry out the usually used mode of unified allocation of resources to each server resource in server.But due to the number of virtual machine
It is relatively large according to measuring, during migrating virtual machine, largely occupancy easily is caused to Internet resources and generates network congestion,
And then the availability of cluster network is impacted, reduce the service quality of cluster network.
It can be seen that a kind of cluster network optimization method is provided, to avoid collecting during migration server virtual machine
Group network generate network congestion, and then guarantee cluster network availability and service quality, be those skilled in the art urgently
It solves the problems, such as.
Summary of the invention
The object of the present invention is to provide a kind of cluster network optimization method, device, equipment and media, in migration server
It avoids cluster network to generate network congestion during virtual machine, and then guarantees the availability and service quality of cluster network.
In order to solve the above technical problems, the present invention provides a kind of cluster network optimization method, comprising:
Obtain the idling-resource information of destination server in cluster network；Wherein, mesh is included at least in idling-resource information
Mark the idle bandwidth of each network link of server；
When the value of the arbitrary target resource items in idling-resource information preset threshold corresponding less than target resource item, sentence
Whether virtual machine has been run in disconnected destination server；
If it is, according to the first logic of propositions in destination server selection target virtual machine, and it is default according to second
Logic selects destination server；
Using idle bandwidth as the constraint of volume of transmitted data, target virtual machine is transmitted to destination server.
Preferably, according to the first logic of propositions in destination server selection target virtual machine, specifically include:
The EMS memory occupation amount of each virtual machine is obtained, and it is total to the occupancy of each regulation resource items to count each virtual machine respectively
With；
The occupancy summation of each virtual machine and the ratio value of EMS memory occupation amount are calculated separately, and selection result is worth maximum void
Intend machine as target virtual machine.
Preferably, destination server is selected according to the second logic of propositions, specifically included:
Target virtual machine is obtained to the resource occupation amount of each regulation resource items, and obtain each server in cluster network can
Use stock number；
The smallest server of ratio value is being chosen between resource occupation amount and available volume of resources in each server as a purpose
Server.
Preferably, cluster network is specially the cluster network of SDN type.
Preferably, target virtual machine is transmitted to destination server specifically:
Target virtual machine is transmitted to destination server by way of multilink parallel transmission.
Preferably, the idling-resource information of destination server in cluster network is obtained specifically:
Idling-resource information is obtained according to prefixed time interval.
Preferably, idling-resource information further includes the CPU idling-resource of destination server and the memory of destination server
Idling-resource.
In addition, the present invention also provides a kind of cluster networks to optimize device, comprising:
Data obtaining module, for obtaining the idling-resource information of destination server in cluster network；Wherein, idling-resource
The idle bandwidth of each network link of destination server is included at least in information；
It is right to be less than target resource item for the value when the arbitrary target resource items in idling-resource information for threshold value judgment module
When the preset threshold answered, judge whether run virtual machine in destination server, if it is, calling selecting module；
Selecting module, for according to the first logic of propositions in destination server selection target virtual machine, and according to second
Logic of propositions selects destination server；
Target virtual machine is transmitted to purpose clothes for the constraint using idle bandwidth as volume of transmitted data by transmission module
Business device.
In addition, the present invention also provides a kind of cluster networks to optimize equipment, comprising:
Memory, for storing computer program；
Processor is realized when for executing computer program such as the step of above-mentioned cluster network optimization method.
In addition, being stored with meter on computer readable storage medium the present invention also provides a kind of computer readable storage medium
Calculation machine program is realized when computer program is executed by processor such as the step of above-mentioned cluster network optimization method.
Cluster network optimization method provided by the present invention, the idling-resource of destination server first in acquisition cluster network
Information, and the idle bandwidth of each network link of the destination server is included at least in the idling-resource information, and then when sky
When the idle degrees value of any one resource in not busy resource information is less than its corresponding preset threshold, then the destination service is determined
Whether virtual machine is run in device, if operation has virtual machine, target is chosen in each virtual machine according to the first logic of propositions
Virtual machine, and the selection destination server according to the second logic of propositions in the cluster network, finally using idle bandwidth as biography
Target virtual machine is transmitted to destination server by the constraint condition of defeated target virtual machine, finally to realize to the excellent of cluster network
Change.As it can be seen that this method is in view of in the server, the data volume of virtual machine is relatively large, holds during migrating virtual machine
The problem of easily causing network congestion, therefore the idle bandwidth of each network link in destination server, Jin Er are obtained in advance
After choosing target virtual machine and destination server, using idle bandwidth as the restriction of transmission objectives virtual machine to destination server,
To ensure when through a certain network link transmission target virtual machine, the data volume of single is no more than the idle band of the network link
Width, the work well of the network link is ensured with this, and then is avoided and caused cluster network to go out during migrating virtual machine
The case where existing network congestion, it ensure that the overall usability and service quality of cluster network.In addition, the present invention also provides one kind
It is same as above that cluster network optimizes device, equipment and medium, beneficial effect.
Detailed description of the invention
In order to illustrate the embodiments of the present invention more clearly, attached drawing needed in the embodiment will be done simply below
It introduces, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ordinary skill people
For member, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of flow chart of cluster network optimization method provided in an embodiment of the present invention；
Fig. 2 is the structure chart that a kind of cluster network provided in an embodiment of the present invention optimizes device.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, rather than whole embodiments.Based on this
Embodiment in invention, those of ordinary skill in the art are without making creative work, obtained every other
Embodiment belongs to the scope of the present invention.
Core of the invention is to provide a kind of cluster network optimization method, to keep away during migration server virtual machine
Exempt from cluster network and generate network congestion, and then guarantees the availability and service quality of cluster network.Another core of the invention
It is to provide a kind of cluster network optimization device, equipment and medium.
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.
Embodiment one
Fig. 1 is a kind of flow chart of cluster network optimization method provided in an embodiment of the present invention.Referring to FIG. 1, cluster net
The specific steps of network optimization method include:
Step S10: the idling-resource information of destination server in cluster network is obtained.
Wherein, the idle bandwidth of each network link of destination server is included at least in idling-resource information.
It is understood that this step is to carry out obtaining for idling-resource information to a certain destination server in cluster network
It takes, the idle degrees information for referring to each resource in the destination server of so-called idling-resource information record.In addition, due to this
There is the case where cluster network congestion, therefore this step is obtained during transmitting data in the focusing on avoiding of method
Idling-resource information in should include at least each network link of the destination server idle bandwidth, idle bandwidth characterization be
Link works normally the data volume that can be carried and transmit under present case, when the data volume by link transmission is beyond the free time
When bandwidth, link may can not work because of congestion, therefore the idle bandwidth reflection of each network link is the link transmission
Ability.
Step S11: when the value of the arbitrary target resource items in idling-resource information is less than, target resource item is corresponding to be preset
When threshold value, judge whether run virtual machine in destination server, if so, thening follow the steps S12.
Step S12: according to the first logic of propositions in destination server selection target virtual machine, and default patrolled according to second
Collect selection destination server.
It should be noted that each target resource Xiang Jun in idling-resource information has corresponding preset threshold, preset threshold
Whether the value for defining each target resource item in idling-resource information is normal, when any one mesh in idling-resource information
When marking the value of resource items less than its corresponding preset threshold, then characterizes the target resource item and be in bottleneck, and then really
Determine the state that destination server belonging to idling-resource information is in overload, the overload of destination server may result in can not be just
Often work to will affect the interaction in the destination server and cluster network between other servers, and then reduces cluster net
The functional reliability of network entirety, therefore need to optimize the destination server in the next steps to ensure cluster where it
The reliability of network.Preset threshold can be set in the form of definite value according to the overall performance and user demand of destination server
It is fixed, it is preferred that be also possible under cluster network between each server, the average value of the target resource item of same type passes through reality
Between each server of Shi Tongji, the average value of the target resource item of same type is capable of the preset threshold that ensures of relative dynamic
Accuracy.
In addition, the first logic of propositions in step S12 in destination server for choosing the destination virtual for needing to migrate
Machine, the second logic of propositions in cluster network choose receive target virtual machine destination server, the first logic of propositions with
Second logic of propositions should be set according to the actual demand of user, if the first logic of propositions is chosen in each virtual machine of destination server
For the most virtual machine of resource occupation amount as target virtual machine, the second logic of propositions is to choose in cluster network idle stock number most
More servers server as a purpose.
Step S13: using idle bandwidth as the constraint of volume of transmitted data, target virtual machine is transmitted to destination server.
It is emphasized that when being migrated to target virtual machine, it should be by step destination server obtained before
Maximum value of the idle bandwidth as transmitted data amount, i.e., can be kept away as the constraint of volume of transmitted data with this using idle bandwidth
Exempt from the idle bandwidth due to volume of transmitted data beyond link and lead to the generation of network congestion, it is ensured that target virtual machine can
Smoothly it is transmitted to destination server.
Cluster network optimization method provided by the present invention, the idling-resource of destination server first in acquisition cluster network
Information, and the idle bandwidth of each network link of the destination server is included at least in the idling-resource information, and then when sky
When the idle degrees value of any one resource in not busy resource information is less than its corresponding preset threshold, then the destination service is determined
Whether virtual machine is run in device, if operation has virtual machine, target is chosen in each virtual machine according to the first logic of propositions
Virtual machine, and the selection destination server according to the second logic of propositions in the cluster network, finally using idle bandwidth as biography
Target virtual machine is transmitted to destination server by the constraint condition of defeated target virtual machine, finally to realize to the excellent of cluster network
Change.As it can be seen that this method is in view of in the server, the data volume of virtual machine is relatively large, holds during migrating virtual machine
The problem of easily causing network congestion, therefore the idle bandwidth of each network link in destination server, Jin Er are obtained in advance
After choosing target virtual machine and destination server, using idle bandwidth as the restriction of transmission objectives virtual machine to destination server,
To ensure when through a certain network link transmission target virtual machine, the data volume of single is no more than the idle band of the network link
Width, the work well of the network link is ensured with this, and then is avoided and caused cluster network to go out during migrating virtual machine
The case where existing network congestion, it ensure that the overall usability and service quality of cluster network.
Embodiment two
On the basis of the above embodiments, the present invention also provides a series of preferred embodiments.
As a preferred embodiment, according to the first logic of propositions in destination server selection target virtual machine,
It specifically includes:
The EMS memory occupation amount of each virtual machine is obtained, and it is total to the occupancy of each regulation resource items to count each virtual machine respectively
With；
The occupancy summation of each virtual machine and the ratio value of EMS memory occupation amount are calculated separately, and selection result is worth maximum void
Intend machine as target virtual machine.
It should be noted that due to consideration that when a certain or a few resource of destination server be more than preset threshold when,
Then this several resources are the current bottleneck of destination server, therefore in order to ensure the normal operation of destination server,
Need to choose destination server after finding the bottleneck of destination server to share the load of destination server.Different from tradition
Greedy algorithm select to occupy selection target of the most target virtual machine of bottleneck as dynamic optimization, present embodiment is selected
Expense brought by dynamic optimization between servers is considered when selecting target virtual machine.In view of expense and virtual machine occupy
Memory size it is related, the memory the big, and the data volume transmitted in dynamic optimization and selection course is more, the system resource of consumption
Also more, therefore, reduce virtual machine and physical host dynamic optimization expense while overload load on host computers to reduce, it should be each
After choosing transfer unit data volume in virtual machine, load capacity reduces the highest virtual machine of degree as target virtual machine.
The ratio value of the occupancy summation and EMS memory occupation amount of signified each virtual machine of calculating in present embodiment, it may be assumed that VS
=(Σ util (resource))/vMemory；Wherein, Σ util (resource) is the occupancy summation of virtual machine,
VMemory is the EMS memory occupation amount of virtual machine, and the value of VS is bigger, then illustrates the load journey that transfer unit data volume can reduce
It spends higher.What present embodiment can be relatively accurate in turn navigates to the target virtual machine for needing to shift, and then opposite improve subtracts
Small object server load overall effect and efficiency.
In addition, as a preferred embodiment, being specifically included according to the second logic of propositions selection destination server:
Target virtual machine is obtained to the resource occupation amount of each regulation resource items, and obtain each server in cluster network can
Use stock number；
The smallest server of ratio value is being chosen between resource occupation amount and available volume of resources in each server as a purpose
Server.
Due to consideration that target virtual machine also will continue to occupy the mesh after target virtual machine is moved to destination server
The related resource of server need ensuring that target virtual machine will not make purpose service therefore when choosing destination server
Device avoids causing bottleneck to threaten the resource in destination server as far as possible, and then in this reality more than on the basis of load threshold
It applies in mode, the smallest server of ratio value will be chosen between resource occupation amount and available volume of resources in each server as a purpose
Server.
In case where providing that resource items are respectively CPU, memory and network bandwidth, the calculation formula of ratio value can have
Body are as follows:
Wherein, PMrcpu, PMrmem, PMrnet points
It Biao Shi not remaining CPU, memory and network bandwidth resources on destination server；VMcpu, VMmem, VMnet respectively indicate mesh
CPU, memory and network bandwidth resources that virtual machine operation needs are marked, each surplus resources of destination server will be greater than void
The demand of quasi- machine is just able to satisfy dynamic optimization and alternative condition, and it is smaller to work as hostcost value, shows target virtual machine and mesh
Server between more match.Present embodiment is opposite to improve the accuracy for choosing destination server, further ensures that whole
The reliability that body cluster network optimizes.
In addition, as a preferred embodiment, cluster network is specially the cluster network of SDN type.
Due to consideration that the network characteristics of SDN type are that control plane separates with data plane, therefore in the collection of SDN
It is more flexible to the acquisition of the resource informations such as network flow for that can ensure under group network basis, therefore based on SDN type
Cluster network realizes the network optimization content of the application, can be improved global optimization efficiency.
In addition, as a preferred embodiment, target virtual machine is transmitted to destination server specifically:
Target virtual machine is transmitted to destination server by way of multilink parallel transmission.
In view of due to recording the idle bandwidth for having each network link of destination server, i.e. target in idling-resource information
Target virtual machine can be transmitted to destination server by a plurality of network link by server, therefore in present embodiment, according to
Constraint of the idle bandwidth of each network link as the corresponding volume of transmitted data of each link, passes through multilink parallel transmission
Target virtual machine is transmitted to destination server by mode, opposite can improve target virtual machine is migrated it is whole to destination server
Body efficiency.
In addition, as a preferred embodiment, obtaining the idling-resource information tool of destination server in cluster network
Body are as follows:
Idling-resource information is obtained according to prefixed time interval.
It should be noted that the prefixed time interval in present embodiment can be depending on the actual demand of user, such as
The action of the cluster network is relatively important, prefixed time interval answer it is relatively short, with this ensure for cluster network optimize
Frequency, it is opposite, if the action importance of the cluster network is general, the value of prefixed time interval can be arranged
It is relatively large, with this avoid frequently because cluster network optimization due to generate corresponding expense.Present embodiment opposite can mention
The mobility of high cluster network optimization, and then improve whole availability.
On the basis of the above embodiment, as a preferred embodiment, idling-resource information further includes target
The CPU idling-resource of server and the memory idling-resource of destination server.
It is understood that in view of cpu resource and memory source are that easily occur the money of resource bottleneck in server
Source Type, thus the idling-resource information in present embodiment in addition to each network link of destination server idle bandwidth with
It outside, further include having the memory idling-resource of the CPU idling-resource and destination server of destination server, and then pass through idle money
The accuracy that source information determines whether the destination server in cluster network needs to optimize is relatively higher, and then can ensure cluster
The network optimization has preferable overall effect.
Embodiment three
Hereinbefore the embodiment of cluster network optimization method is described in detail, the present invention also provides one kind
Cluster network corresponding with this method optimizes device, since the embodiment of device part and the embodiment of method part are mutually right
It answers, therefore the embodiment of device part refers to the description of the embodiment of method part, wouldn't repeat here.
Fig. 2 is the structure chart that a kind of cluster network provided in an embodiment of the present invention optimizes device.The embodiment of the present invention provides
Cluster network optimize device, comprising:
Data obtaining module 10, for obtaining the idling-resource information of destination server in cluster network；Wherein, idle money
The idle bandwidth of each network link of destination server is included at least in source information.
Threshold value judgment module 11 is less than target resource item for the value when the arbitrary target resource items in idling-resource information
When corresponding preset threshold, judge whether run virtual machine in destination server, if it is, calling selecting module 12.
Selecting module 12, for according to the first logic of propositions in destination server selection target virtual machine, and according to the
Two logic of propositions select destination server.
Target virtual machine is transmitted to purpose for the constraint using idle bandwidth as volume of transmitted data by transmission module 13
Server.
Cluster network provided by the present invention optimizes device, the idling-resource of destination server first in acquisition cluster network
Information, and the idle bandwidth of each network link of the destination server is included at least in the idling-resource information, and then when sky
When the idle degrees value of any one resource in not busy resource information is less than its corresponding preset threshold, then the destination service is determined
Whether virtual machine is run in device, if operation has virtual machine, target is chosen in each virtual machine according to the first logic of propositions
Virtual machine, and the selection destination server according to the second logic of propositions in the cluster network, finally using idle bandwidth as biography
Target virtual machine is transmitted to destination server by the constraint condition of defeated target virtual machine, finally to realize to the excellent of cluster network
Change.As it can be seen that the present apparatus is in view of in the server, the data volume of virtual machine is relatively large, holds during migrating virtual machine
The problem of easily causing network congestion, therefore the idle bandwidth of each network link in destination server, Jin Er are obtained in advance
After choosing target virtual machine and destination server, using idle bandwidth as the restriction of transmission objectives virtual machine to destination server,
To ensure when through a certain network link transmission target virtual machine, the data volume of single is no more than the idle band of the network link
Width, the work well of the network link is ensured with this, and then is avoided and caused cluster network to go out during migrating virtual machine
The case where existing network congestion, it ensure that the overall usability and service quality of cluster network.
Example IV
The present invention also provides a kind of cluster networks to optimize equipment, comprising:
Memory, for storing computer program；
Processor is realized when for executing computer program such as the step of above-mentioned cluster network optimization method.
Cluster network provided by the present invention optimizes equipment, the idling-resource of destination server first in acquisition cluster network
Information, and the idle bandwidth of each network link of the destination server is included at least in the idling-resource information, and then when sky
When the idle degrees value of any one resource in not busy resource information is less than its corresponding preset threshold, then the destination service is determined
Whether virtual machine is run in device, if operation has virtual machine, target is chosen in each virtual machine according to the first logic of propositions
Virtual machine, and the selection destination server according to the second logic of propositions in the cluster network, finally using idle bandwidth as biography
Target virtual machine is transmitted to destination server by the constraint condition of defeated target virtual machine, finally to realize to the excellent of cluster network
Change.As it can be seen that this equipment is in view of in the server, the data volume of virtual machine is relatively large, holds during migrating virtual machine
The problem of easily causing network congestion, therefore the idle bandwidth of each network link in destination server, Jin Er are obtained in advance
After choosing target virtual machine and destination server, using idle bandwidth as the restriction of transmission objectives virtual machine to destination server,
To ensure when through a certain network link transmission target virtual machine, the data volume of single is no more than the idle band of the network link
Width, the work well of the network link is ensured with this, and then is avoided and caused cluster network to go out during migrating virtual machine
The case where existing network congestion, it ensure that the overall usability and service quality of cluster network.
In addition, being stored with meter on computer readable storage medium the present invention also provides a kind of computer readable storage medium
Calculation machine program is realized when computer program is executed by processor such as the step of above-mentioned cluster network optimization method.
Computer readable storage medium provided by the present invention, the idle money of destination server first in acquisition cluster network
Source information, and the idle bandwidth of each network link of the destination server is included at least in the idling-resource information, and then work as
When the idle degrees value of any one resource in idling-resource information is less than its corresponding preset threshold, then determine that the target takes
Whether virtual machine is run in business device, if operation has virtual machine, mesh is chosen in each virtual machine according to the first logic of propositions
Mark virtual machine, and the selection destination server according to the second logic of propositions in the cluster network, finally using idle bandwidth as
Target virtual machine is transmitted to destination server by the constraint condition of transmission objectives virtual machine, finally to realize to cluster network
Optimization.As it can be seen that this computer readable storage medium is in view of in the server, the data volume of virtual machine is relatively large, is migrating
The problem of network congestion is be easy to cause during virtual machine, therefore each network link is obtained in destination server in advance
Idle bandwidth, and then after choosing target virtual machine and destination server, using idle bandwidth as transmission objectives virtual machine to mesh
Server restriction, to ensure when through a certain network link transmission target virtual machine, the data volume of single is no more than should
The idle bandwidth of network link, the work well of the network link is ensured with this, and then avoids the process in migration virtual machine
In cause cluster network the case where network congestion occur, ensure that the overall usability and service quality of cluster network.
Detailed Jie has been carried out to a kind of cluster network optimization method provided by the present invention, device, equipment and medium above
It continues.Each embodiment is described in a progressive manner in specification, the highlights of each of the examples are with other embodiments
Difference, the same or similar parts in each embodiment may refer to each other.For device, equipment disclosed in embodiment and
For medium, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method portion
It defends oneself bright.It should be pointed out that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, it can be with several improvements and modifications are made to the present invention, these improvement and modification also fall into the protection of the claims in the present invention
In range.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Claims (10)
1. a kind of cluster network optimization method characterized by comprising
Obtain the idling-resource information of destination server in cluster network；Wherein, institute is included at least in the idling-resource information
State the idle bandwidth of each network link of destination server；
When the value of the arbitrary target resource items in the idling-resource information is less than the corresponding preset threshold of the target resource item
When, judge whether run virtual machine in the destination server；
If it is, according to the first logic of propositions in the destination server selection target virtual machine, and it is default according to second
Logic selects destination server；
Using the idle bandwidth as the constraint of volume of transmitted data, the target virtual machine is transmitted to the destination server.
2. the method according to claim 1, wherein it is described according to the first logic of propositions in the destination server
Middle selection target virtual machine, specifically includes:
The EMS memory occupation amount of each virtual machine is obtained, and counts each virtual machine respectively to the occupancy of each regulation resource items
Summation；
Calculate separately the occupancy summation of each virtual machine and the ratio value of the EMS memory occupation amount, and selection result value
The maximum virtual machine is as the target virtual machine.
3. according to the method described in claim 2, it is characterized in that, it is described according to the second logic of propositions select destination server,
It specifically includes:
The target virtual machine is obtained to the resource occupation amount of each regulation resource items, and obtains and is respectively taken in the cluster network
The available volume of resources of business device；
The smallest clothes of ratio value between the resource occupation amount and the available volume of resources are being chosen in each server
Device be engaged in as the destination server.
4. the method according to claim 1, wherein the cluster network is specially the cluster network of SDN type.
5. the method according to claim 1, wherein described be transmitted to the purpose clothes for the target virtual machine
Business implement body are as follows:
The target virtual machine is transmitted to the destination server by way of multilink parallel transmission.
6. the method according to claim 1, wherein the idle money for obtaining destination server in cluster network
Source information specifically:
The idling-resource information is obtained according to prefixed time interval.
7. method described in -6 any one according to claim 1, which is characterized in that the idling-resource information further includes described
The CPU idling-resource of destination server and the memory idling-resource of the destination server.
8. a kind of cluster network optimizes device characterized by comprising
Data obtaining module, for obtaining the idling-resource information of destination server in cluster network；Wherein, the idling-resource
The idle bandwidth of each network link of the destination server is included at least in information；
Threshold value judgment module is less than the target resource for the value when the arbitrary target resource items in the idling-resource information
When the corresponding preset threshold of item, judge whether run virtual machine in the destination server, if it is, calling selection mould
Block；
The selecting module, for according to the first logic of propositions in the destination server selection target virtual machine, and according to
Second logic of propositions selects destination server；
The target virtual machine is transmitted to institute for the constraint using the idle bandwidth as volume of transmitted data by transmission module
State destination server.
9. a kind of cluster network optimizes equipment characterized by comprising
Memory, for storing computer program；
Processor realizes cluster network optimization as described in any one of claim 1 to 7 when for executing the computer program
The step of method.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program, the computer program realize cluster network optimization side as described in any one of claim 1 to 7 when being executed by processor
The step of method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810921913.5A CN109039933B (en) | 2018-08-14 | 2018-08-14 | Cluster network optimization method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810921913.5A CN109039933B (en) | 2018-08-14 | 2018-08-14 | Cluster network optimization method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109039933A true CN109039933A (en) | 2018-12-18 |
| CN109039933B CN109039933B (en) | 2022-07-08 |
Family
ID=64634038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810921913.5A Active CN109039933B (en) | 2018-08-14 | 2018-08-14 | Cluster network optimization method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109039933B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110225033A (en) * | 2019-06-11 | 2019-09-10 | 西安电子科技大学 | Isomery controller collection group business based on abnormality sensing actively changes system and method |
| WO2020029994A1 (en) * | 2018-08-07 | 2020-02-13 | 中兴通讯股份有限公司 | Virtual pc resource sharing method, device, computer device and storage medium |
| CN111444008A (en) * | 2018-12-29 | 2020-07-24 | 北京奇虎科技有限公司 | Inter-cluster service migration method and device |
| CN114363328A (en) * | 2021-11-19 | 2022-04-15 | 三维通信股份有限公司 | Task coordination method and system for server cluster |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060069761A1 (en) * | 2004-09-14 | 2006-03-30 | Dell Products L.P. | System and method for load balancing virtual machines in a computer network |
| CN103888501A (en) * | 2012-12-24 | 2014-06-25 | 华为技术有限公司 | Virtual machine migration method and device |
| CN106528270A (en) * | 2016-11-16 | 2017-03-22 | 航天信息股份有限公司 | Automatic migration method and system of virtual machine based on OpenStack cloud platform |
-
2018
- 2018-08-14 CN CN201810921913.5A patent/CN109039933B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060069761A1 (en) * | 2004-09-14 | 2006-03-30 | Dell Products L.P. | System and method for load balancing virtual machines in a computer network |
| CN103888501A (en) * | 2012-12-24 | 2014-06-25 | 华为技术有限公司 | Virtual machine migration method and device |
| CN106528270A (en) * | 2016-11-16 | 2017-03-22 | 航天信息股份有限公司 | Automatic migration method and system of virtual machine based on OpenStack cloud platform |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020029994A1 (en) * | 2018-08-07 | 2020-02-13 | 中兴通讯股份有限公司 | Virtual pc resource sharing method, device, computer device and storage medium |
| CN111444008A (en) * | 2018-12-29 | 2020-07-24 | 北京奇虎科技有限公司 | Inter-cluster service migration method and device |
| CN111444008B (en) * | 2018-12-29 | 2024-04-16 | 北京奇虎科技有限公司 | Inter-cluster service migration method and device |
| CN110225033A (en) * | 2019-06-11 | 2019-09-10 | 西安电子科技大学 | Isomery controller collection group business based on abnormality sensing actively changes system and method |
| CN110225033B (en) * | 2019-06-11 | 2021-05-18 | 西安电子科技大学 | Active migration system and method for heterogeneous controller cluster service based on abnormal perception |
| CN114363328A (en) * | 2021-11-19 | 2022-04-15 | 三维通信股份有限公司 | Task coordination method and system for server cluster |
| CN114363328B (en) * | 2021-11-19 | 2024-02-20 | 三维通信股份有限公司 | Task collaboration method and system for server cluster |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109039933B (en) | 2022-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109039933A (en) | A kind of cluster network optimization method, device, equipment and medium | |
| CN112153700B (en) | Network slice resource management method and equipment | |
| CN106533723B (en) | Virtual resource scheduling method, device and system | |
| Chaczko et al. | Availability and load balancing in cloud computing | |
| CN103338252B (en) | Realizing method of distributed database concurrence storage virtual request mechanism | |
| Dong et al. | Energy-performance tradeoffs in IaaS cloud with virtual machine scheduling | |
| US20130326038A1 (en) | Management of datacenters for fault tolerance and bandwidth | |
| CN108196788A (en) | QoS index monitoring method, device, storage medium | |
| CN113348651B (en) | Dynamic inter-cloud placement of sliced virtual network functions | |
| CN109976917A (en) | A kind of load dispatching method, device, load dispatcher, storage medium and system | |
| CN102111337A (en) | Method and system for task scheduling | |
| CN104539744B (en) | A kind of the media edge cloud dispatching method and device of two benches cooperation | |
| CN108509157A (en) | A kind of data balancing method and device applied to distributed file system | |
| CN108196935A (en) | A kind of energy saving moving method of virtual machine towards cloud computing | |
| EP4170491A1 (en) | Resource scheduling method and apparatus, electronic device, and computer-readable storage medium | |
| CN107291539A (en) | Cluster program scheduler method based on resource significance level | |
| CN109491788A (en) | A kind of virtual platform implementation of load balancing and device | |
| CN113300861B (en) | Network slice configuration method, device and storage medium | |
| Panda et al. | User allocation in mobile edge computing: A deep reinforcement learning approach | |
| CN110597598B (en) | Control method for virtual machine migration in cloud environment | |
| Abdullah et al. | A reliable, TOPSIS-based multi-criteria, and hierarchical load balancing method for computational grid | |
| CN110471761A (en) | Control method, user equipment, storage medium and the device of server | |
| CN106020732A (en) | Node disk space determining method and system | |
| CN101853185B (en) | Blade server and service dispatching method thereof | |
| US20160006617A1 (en) | Cloud application bandwidth modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |