CN108306912A - Virtual network function management method and its device, network function virtualization system - Google Patents
Virtual network function management method and its device, network function virtualization system Download PDFInfo
- Publication number
- CN108306912A CN108306912A CN201710022734.3A CN201710022734A CN108306912A CN 108306912 A CN108306912 A CN 108306912A CN 201710022734 A CN201710022734 A CN 201710022734A CN 108306912 A CN108306912 A CN 108306912A
- Authority
- CN
- China
- Prior art keywords
- host
- network function
- virtual network
- key performance
- performance indicator
- 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.)
- Withdrawn
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1029—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers using data related to the state of servers by a load balancer
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
A kind of virtual network function management method of offer of the embodiment of the present invention and its device, network function virtualization system, by the KPI Key Performance Indicator for obtaining each host in mainframe cluster, according to the KPI Key Performance Indicator of each host and preset load balancing, management strategy is generated, the virtual network function in each host is managed according to the management strategy of generation；Realize the resources balance distribution of network function virtualization, it is excessive to avoid the occurrence of part host operating load, traffic affecting treatment effeciency, improves the utilization rate and network function virtualization high availability and stability of virtual network function resource.
Description
Technical field
The present invention relates to network function virtualization field more particularly to a kind of virtual network function (Virtual Network
Functions, VNF) management method and its device, network function virtualize (Network Function
Virtualization, NFV) system.
Background technology
Currently, in traditional telecom market, the network element of different manufacturers different business often operate in different hardware devices,
On operating system, part is even using some proprietary hardware.In different business scopes and different periods, if
Standby utilization rate is not quite similar, and the configuration of these equipment needs the peak load requirements for meeting single business, but when most of
Time is limited these resources due to the demand of practical business and cannot be utilized well, to cause the resource wave of equipment
Take.
In recent years, cloud computing, virtualization technology are quickly grown, and bring many innovations, while also being brought very to operator
Big pressure, operator, which faces, finds new income increase point, to offset OTT (Video service based on open the Internet) business
The influence brought, while OPEX (operation cost) is reduced, it quickly commences business, it is therefore desirable to as follows by virtualizing emphasis solution
Railway Project：
(1) the United Dispatching problem of resource；
(2) network planning can not quickly, be flexibly adjusted；
(3) buying and operation cost are reduced；
(4) professional ability exploitation is difficult；
Gradually ripe and perfect with virtualization standard, the rapid deployment of network, flexible adjustment become possible, NFV
Technology is come into being.The Its Relevant Technology Standards of NFV define the technology that conventional telecommunications network element is migrated to virtualization cloud environment and realize.
On cloud environment, Telecommunication Network Element is no longer that proprietary hardware carries software unit one by one, but general VDU (dummy sections
Administration's unit) telecommunication service software unit is carry, virtual network function VNF is constituted, by VNF network consisting services (Network
Service, NS).
Cloud environment and its virtual machine constitute the Virtual base facility layer NFVI (NFV infrastructure) of VNF, provide calculating money
Source pond, memory resource pool, Internet resources pond, but by virtualization technology, again without the equiblibrium mass distribution for realizing device resource,
The utilization rate of device resource or relatively low.Therefore, VNF how being distributed in virtualization pool according to load balancing, to
Uniformly share VNF loads in entire NFVI levels, the high availability and stability of NFV will be directly influenced.
Invention content
Virtual network function management method provided in an embodiment of the present invention and its device, network function virtualization system, it is main
Technical problems to be solved are to realize the resources balance distribution of network function virtualization, improve network function and virtualize high availability
And the problem of stability.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of virtual network function management method, including：
Obtain the KPI Key Performance Indicator of each host in mainframe cluster；
According to the KPI Key Performance Indicator of each host and preset load balancing, management strategy is generated；
The virtual network function in each host is managed according to the management strategy.
The embodiment of the present invention also provides a kind of virtual network function managing device, including：Decision-making module and management module,
In：
The decision-making module is used to obtain the KPI Key Performance Indicator of each host in mainframe cluster；According to the pass of each host
Key performance indicators and preset load balancing generate management strategy；
The management module is used to manage the virtual network function in each host according to the management strategy.
The embodiment of the present invention also provides a kind of network function virtualization system, including：Network function virtualize coordinator and
Virtual network function manager and above-mentioned virtual network function managing device, the virtual network function managing device are used
In the corresponding virtual network function of the management virtual network function manager.
The embodiment of the present invention also provides a kind of computer storage media, and computer is stored in the computer storage media
Executable instruction, the computer executable instructions are for executing virtual network function management method above-mentioned.
The beneficial effects of the invention are as follows：
The virtual network function management method and its device that there is provided according to embodiments of the present invention, network function virtualization system
And computer storage media, by obtaining the KPI Key Performance Indicator of each host in mainframe cluster, according to the crucial achievement of each host
Index and preset load balancing are imitated, management strategy is generated, according to the management strategy of generation to virtual in each host
Network function is managed；It realizes and mainframe cluster is monitored in real time by the KPI Key Performance Indicator of each host, avoid out
Now host operating load in part is excessive, traffic affecting treatment effeciency, according to preset load balancing to the void of each host
Quasi- network function carries out balanced management, realizes the United Dispatching of resource, solves the problems, such as device resource waste, greatly improve
The utilization rate of virtual network function resource, while also improving the high availability and stability of network function virtualization.
Description of the drawings
Fig. 1 is a kind of process chart for virtual network function management method that the embodiment of the present invention one provides；
Fig. 2 is another flow chart of virtual network function management method provided by Embodiment 2 of the present invention；
Fig. 3 screens move out host, the VNF in host of moving out, the processing for moving into host to be provided by Embodiment 2 of the present invention
Flow chart；
Fig. 4 is a kind of virtual network function managing device structural schematic diagram that the embodiment of the present invention three provides；
Fig. 5 is a kind of network function virtualization system structural schematic diagram that the embodiment of the present invention four provides.
Specific implementation mode
The embodiment of the present invention is described in further detail below by specific implementation mode combination attached drawing.
Embodiment one：
In the prior art, x86 servers, storage and the exchange that VNF uses professional standard are virtualized by network function
Equipment replaces the privately owned dedicated network element device of some communication networks, but there is no the equiblibrium mass distribution for realizing resource so that sets
The utilization rate of standby resource or relatively low.To solve the above-mentioned problems, an embodiment of the present invention provides a kind of virtual network functions
Management method is monitored by the KPI Key Performance Indicator to each host in mainframe cluster, and according to the Key Performance of each host
Index and preset load balancing generate corresponding management strategy and carry out equilibrium to the virtual network function in each host
Management, to improve the utilization rate of virtual network function resource, solution is led since the distribution of resource is unbalanced in the prior art
The problem of causing resource serious waste.
Fig. 1 is referred to, Fig. 1 is a kind of process chart of virtual network function management method provided in this embodiment, should
Method includes：
S101 obtains the KPI Key Performance Indicator of each host in mainframe cluster, and the KPI Key Performance Indicator includes going through for host
The average value of history load, present load and load.
In the present embodiment, the KPI Key Performance Indicator for obtaining each host in mainframe cluster is regular particular by VIM
Detection acquires, which includes：
Between when detecting in section, the load data of each host in mainframe cluster is periodically obtained；
According to the load data for getting each host, the KPI Key Performance Indicator of each host is calculated.
Specifically, the load data includes historic load and current real time load data, VIM can periodically be led to detection
The load data of each host in machine cluster, and the average value of the load data of each host in each detection time section is calculated, according to institute
The KPI Key Performance Indicator of mean value calculation corresponding host is stated, or calculates the pass of corresponding host according to the load data of each host
Key performance indicators counts the KPI Key Performance Indicator of each host in current sensing time, then calculates the flat of KPI Key Performance Indicator
Mean value.
In the present embodiment, it when the load data average value of each host of acquisition reaches preset load threshold, then touches
It sends out and the balanced of the virtual network function of host is adjusted, or when the average value of KPI Key Performance Indicator reaches preset Key Performance
When metrics-thresholds, the equilibrium adjustment to the virtual network function of host is also triggered.
In the present embodiment, the KPI Key Performance Indicator for obtaining each host in mainframe cluster can also be by judging dynamic
Whether the balanced period reaches；When reaching in the preset dynamic equalization period, the automatic load number for obtaining each host in mainframe cluster
According to according to the load data of each host, the KPI Key Performance Indicator of each host is calculated.
Specifically, when needing to carry out dynamic equalization, reached when the preset dynamic equalization period, then automatic trigger once obtains
The load data of each host is taken, and calculates corresponding KPI Key Performance Indicator, judges whether to reach preset load data threshold value, if
Reach, then the processing of moving out for carrying out load data is not processed if being not up to.After processing is completed, restorer is normal
Operating status continues the arrival for detecting next dynamic equalization period.
S102 generates management strategy according to the KPI Key Performance Indicator of each host and preset load balancing.
In the present embodiment, which refers specifically to the load data of host being more than preset load number
It carries out loading processing of moving out according to the load data on the corresponding host of threshold value, preferential moves out to the lower host of load data
On, it is used to realize the balanced of resource.
In the present embodiment, the process for generating management strategy further includes specifically：
Whether adjustment threshold value is more than to the KPI Key Performance Indicator of each host respectively, when KPI Key Performance Indicator is flat for load data
When mean value, corresponding adjusted value is preset load data threshold value.
If the KPI Key Performance Indicator of the first host is more than adjustment threshold value, using the first host as host of moving out.
The KPI Key Performance Indicator of each host is ranked up, selected and sorted is in last finger from all KPI Key Performance Indicators
For second host of fixed number amount as host is moved into, it can be multiple that this, which moves into host, and each hosts of the N of selected and sorted rearward, which are used as, to be moved
Enter host, the load data polarity moved out selection is specifically needed according to host of moving out.
According to the KPI Key Performance Indicator and preset load balancing of the first host and the second host, generate corresponding
Management strategy, the management strategy are moved out to the second host for the virtual network function in the host that will move out.
In the present embodiment, the KPI Key Performance Indicator according to each host and preset load balancing, it is raw
Further include at management strategy：Obtain the load data of the virtual deployment unit of each virtual network function in host of moving out；
According to the load data of the virtual deployment unit of each virtual network function, virtual network function is ranked up；
According to ranking results, select the virtual network function beyond adjustment threshold value as migration target；
According to the host of moving out, the migration target and it is described move into host, generate management strategy；The management strategy
For the virtual network function as migration target in the host that will move out, moves to and described move into host.
S103 manages the virtual network function in each host according to management strategy.
In the present embodiment, before step S102 generates management strategy, further include：
Budget KPI Key Performance Indicator of the host after moving into the migration target is moved into described in budget；
Judge whether the budget KPI Key Performance Indicator is more than the adjustment threshold value；
If so, host is moved into gravity treatment, and correct the management strategy.
After step S102 generates management strategy, further include：
Judge whether occur being not present host of moving out, or appearance is in the presence of host of moving out, there is no the judgements for moving into host
As a result；
If there is the judging result, the management strategy is not generated.
Virtual network function management method provided in an embodiment of the present invention, by the key for obtaining each host in mainframe cluster
Performance indicators generates management strategy, according to life according to the KPI Key Performance Indicator of each host and preset load balancing
At management strategy the virtual network function in each host is managed；Realize the resources balance point of network function virtualization
Cloth improves network function virtualization high availability and stability.
Further, it is super for KPI Key Performance Indicator when the load data of each host is monitored in mainframe cluster
When crossing adjustment threshold values, management strategy is automatically generated, the virtual network function in host of moving out is moved out to moving according to management strategy
Enter in host, the equiblibrium mass distribution of network function virtualization load is furthermore achieved, improves the profit of virtual network function resource
With rate.
Embodiment two：
Referring to FIG. 2, Fig. 2 is another flow chart of virtual network function management method provided in this embodiment.This implementation
The mode that example provides is applicable in the processing of general and special VNF Autonomic Migration Frameworks under virtualization scene, and processing step specifically includes：
S201 enables cloud environment mainframe cluster VNF load balancings.
S202, the load data of taken at regular intervals VNF hosts.
S203 preserves load data.
S204 calculates the average value of historic load, specifically periodically calculates the average value of the historic load of each host
As KPI Key Performance Indicator, when the threshold value defined in it is more than load balancing, triggering migration warming-up exercise.
S205 carries out load balancing migration strategy information conditional and judges according to the KPI Key Performance Indicator information of each host,
If being no more than threshold values, periodically go to step S202, otherwise continues.
S206 is screened host of moving out, the VNF in host of moving out, is moved into host by load-balancing algorithm.
In this step, the host that had more than load threshold is specifically filtered out from the All hosts of mainframe cluster is made
For host of moving out, screening loads minimum multiple main frames and alternately moves into host, determines to need to move out from which by calculating
Which which VNF that moves out on host move into host to.
S207 implements the VNF migrations to host of moving out one by one.
S208 updates the VNF configuration informations of each host in mainframe cluster.
In the present embodiment, the VNF in move out host, host of moving out is screened for step S206, moves into the specific of host
Screening process is as shown in figure 3, the step specifically includes：
S301, historic load average value of the All hosts in observation cycle in computing cluster.
S302 sorts by historic load average value size.
S303, load are host of moving out more than the host acknowledgement of threshold value.
S304 calculates historic load average values of the VNF to move out on host in observation cycle.
S305, load data from low to high, select host alternately to move into host.
S306 screens host of moving out, the VNF moved into, host of alternately moving out according to load-balancing algorithm.
S307, acquisition is currently optimal to move into host.
S308 obtains current optimal host of moving out.
S309, host is moved into judgement and whether host of moving out is empty, if one of them is sky, this migration plan is complete
At otherwise steering step S315 turns to step S310.
S310 is calculated and is attempted to move to the VNF that override is moved out corresponding to the load beyond threshold portion on host
It is prepreerence to move on host.
S311 confirms this migration plan if moving into load on host computers data is no more than threshold value, turns to step S312, otherwise
Turn to step 313.
S312 confirms migration plan, and stores migration plan.
S313 removes host of currently moving out, and continues calculating second priority and moves into host.
S314, will complete the host removal of moving out of migration plan, and places of going forward side by side preferentially is moved out the calculating of host.
S315 obtains complete migration plan.
In the present embodiment, further include the operation information record log according to above-mentioned migration after the completion of migrating VNF, into
Row preserve, facilitate it is subsequent move back or restores reset condition reservation Query Information.
Embodiment three：
Referring to FIG. 4, Fig. 4 is a kind of virtual network function managing device structural schematic diagram provided in this embodiment.The void
Intending network function managing device 4 includes：Decision-making module 41 and management module 42, wherein：
Decision-making module 41 is used to obtain the KPI Key Performance Indicator of each host in mainframe cluster, and according to the pass of each host
Key performance indicators and preset load balancing generate management strategy.
In the present embodiment, the Key Performance that the real decision-making module 41 obtains each host particular by following manner refers to
Mark：
Between when detecting in section, the load data of each host in mainframe cluster is periodically obtained, according to getting each master
The KPI Key Performance Indicator of each host is calculated in the load data of machine, specifically, the load data include historic load and
Current real time load data.
Specifically, decision-making module 41 passes through virtual deployment unit used in host in VIM taken at regular intervals clusters and VNF
Load data, the load data average value of each host is calculated according to the load data or corresponding Key Performance refers to
Mark.
In the present embodiment, the decision-making module 41 can also be according to judging whether the dynamic equalization period reaches, described
When the dynamic equalization period reaches, the load data of each host in mainframe cluster is obtained, according to the load data of each host, meter
Calculation obtains the KPI Key Performance Indicator of each host.
Specifically, when needing to carry out dynamic equalization, reached when the preset dynamic equalization period, then automatic trigger once obtains
The load data of each host is taken, and calculates corresponding KPI Key Performance Indicator, judges whether to reach preset load data threshold value, if
Reach, then the processing of moving out for carrying out load data is not processed if being not up to.After processing is completed, restorer is normal
Operating status continues the arrival for detecting next dynamic equalization period.
When the load data average value of each host of acquisition reaches preset load threshold, then triggering is to the virtual of host
The balanced adjustment of network function, or when the average value of KPI Key Performance Indicator reaches preset KPI Key Performance Indicator threshold value,
It triggers and the balanced of the virtual network function of host is adjusted.
In the present embodiment, load balancing includes adjustment threshold value, and decision-making module 41 for judging each host respectively
Whether KPI Key Performance Indicator is more than adjustment threshold value；If the KPI Key Performance Indicator of the first host is more than adjustment threshold value, by the first master
Machine is as host of moving out；It is sorted according to the KPI Key Performance Indicator of each host, the specified number for selecting KPI Key Performance Indicator sequence last
Second host of amount, as moving into host；Host and host is moved into according to moving out, and generates management strategy, management strategy is used for will
The virtual network function moved out in host, moves to and moves into host.
Management module 42 is used to manage the virtual network function in each host according to management strategy.
In the present embodiment, the specific void for being additionally operable to obtain each virtual network function in host of moving out of the decision-making module 41
The load data of quasi- deployment unit；According to the load data of the virtual deployment unit of each virtual network function, to virtual network work(
It can be ranked up；According to ranking results, select the virtual network function beyond adjustment threshold value as migration target；According to the master that moves out
Machine migrates target and moves into host, generates management strategy；Management strategy is used to move out in host as migrating the virtual of target
Network function moves to and moves into host.
In the present embodiment, the decision-making module 41 is additionally operable to budget and moves into budget of the host after moving into migration target
KPI Key Performance Indicator；Judge whether budget KPI Key Performance Indicator is more than adjustment threshold value；If so, host is moved into gravity treatment, and correct
Management strategy.
In the present embodiment, it after the decision-making module 41 generates management strategy, is additionally operable to judge whether to occur not deposit
In host of moving out, or appearance is in the presence of host of moving out, there is no the judging results for moving into host；If there is the judging result,
The management strategy is not generated then.
Virtual network function managing device provided in this embodiment obtains each host in mainframe cluster by decision-making module
KPI Key Performance Indicator, and according to the KPI Key Performance Indicator of each host and preset load balancing, generate management strategy, pipe
The management strategy that reason module is generated according to decision-making module carries out balanced management to the virtual network function in each host；Realize network
The resources balance of virtualization of function is distributed, and improves network function virtualization high availability and stability.
Further, by the way that the threshold value of load data or the threshold value of KPI Key Performance Indicator is arranged, judgement gets each master
Whether the load data of machine or the KPI Key Performance Indicator being calculated are more than threshold value, and dynamic equalization is controlled by the comparison of threshold value
Adjustment is furthermore achieved the equiblibrium mass distribution of network function virtualization load, improves the utilization rate of virtual network function resource,
Simultaneously the host that also avoids by load is excessive and it is huge to cause the damage of equipment and traffic affecting processing institute to bring to user
Big loss.
Example IV：
Fig. 5 is referred to, Fig. 5 is network function virtualization system structural schematic diagram provided in this embodiment.The present embodiment carries
The network function virtualization system 5 of confession includes：Network function virtualizes coordinator (NFV Orchestrator, NFV coordinator)
51 and virtual network function manager (VNF Manager, VNF manager) 52 and above-mentioned virtual network function management dress
4 are set, virtual network function managing device 4 is for managing 52 corresponding virtual network function of the virtual network function manager.
In the present embodiment, virtual network function managing device 4 can be used as autonomous device, and setting is in network function void
Except quasi-ization coordinator 51 and virtual network function manager 52；Or the virtual network function managing device 4 is arranged in institute
It states inside network function virtualization coordinator 51；Alternatively, the setting of decision-making module 41 of the virtual network function managing device 4 exists
Inside the network function virtualization coordinator 52, the management module 42 of the virtual network function managing device 4 is arranged in institute
It states inside virtual network function manager 52.
Preferably, in the present embodiment, network function virtualization system 5 by NFV coordinators built-in VNF from moving
The decision-making module 41 of shifting monitors relevant KPI Key Performance Indicator in each host, when judging that KPI Key Performance Indicator reaches preset valve
When value, relevant load data balance policy is issued to the corresponding VNF managers of the VNF, is moved out to needs by VNF managers
VNF carries out processing of moving out.
Further, the network function virtualization system 5 is when judging that KPI Key Performance Indicator reaches preset threshold values, also
Can by holding operation according to the relevant reducing and expansion of preset strategy execution, be the larger Host Extensions running memory of load data or
Person's buffer area.
In the present embodiment, the KPI Key Performance Indicator of each host in mainframe cluster is obtained by the decision-making module 41, and
According to the KPI Key Performance Indicator of each host and preset load balancing, management strategy is generated.
Specifically, decision-making module 41 when detecting between in section, periodically obtain the load number of each host in mainframe cluster
According to according to the load data for getting each host, the KPI Key Performance Indicator of each host being calculated, specifically, the load data
Including historic load and current real time load data.
Management module 42 manages the virtual network function in each host according to management strategy.
In conclusion virtual network function management method provided by the invention and its device, network function virtualization system,
By obtaining the KPI Key Performance Indicator of each host in mainframe cluster, according to the KPI Key Performance Indicator of each host and preset negative
Balance policy is carried, management strategy is generated, the virtual network function in each host is managed according to the management strategy of generation；It is real
The resources balance distribution of existing network function virtualization, it is excessive to avoid the occurrence of part host operating load, traffic affecting processing effect
Rate, the utilization rate and network function for improving virtual network function resource virtualize high availability and stability.
Obviously, those skilled in the art should be understood that each module of the embodiments of the present invention or each step can be used
General computing device realizes that they can be concentrated on a single computing device, or be distributed in multiple computing device institutes
On the network of composition, optionally, they can be realized with the program code that computing device can perform, it is thus possible to by them
It is stored in computer storage media (ROM/RAM, magnetic disc, CD) and is performed by computing device, and in some cases, it can
With the steps shown or described are performed in an order that is different from the one herein, or they are fabricated to each integrated circuit dies
Block, or by them multiple modules or step be fabricated to single integrated circuit module and realize.So the present invention does not limit
It is combined in any specific hardware and software.
It cannot recognize being further described made by the embodiment of the present invention the above content is specific embodiment is combined
The specific implementation of the fixed present invention is confined to these explanations.For those of ordinary skill in the art to which the present invention belongs,
Without departing from the inventive concept of the premise, a number of simple deductions or replacements can also be made, all shall be regarded as belonging to the present invention
Protection domain.
Claims (14)
1. a kind of virtual network function management method, including：
Obtain the KPI Key Performance Indicator of each host in mainframe cluster；
According to the KPI Key Performance Indicator of each host and preset load balancing, management strategy is generated；
The virtual network function in each host is managed according to the management strategy.
2. virtual network function management method according to claim 1, which is characterized in that each in the acquisition mainframe cluster
The KPI Key Performance Indicator of host includes：
Between when detecting in section, the load data of each host in mainframe cluster is periodically obtained；
According to the load data of each host, the KPI Key Performance Indicator of each host is calculated.
3. virtual network function management method according to claim 1, which is characterized in that each in the acquisition mainframe cluster
The KPI Key Performance Indicator of host includes：
Judge whether the dynamic equalization period reaches；
When reaching in the dynamic equalization period, the load data of each host in mainframe cluster is obtained；
According to the load data of each host, the KPI Key Performance Indicator of each host is calculated.
4. virtual network function management method according to any one of claims 1 to 3, which is characterized in that the load is equal
Weighing apparatus strategy includes adjustment threshold value, the KPI Key Performance Indicator according to each host and preset load balancing, is generated
Management strategy includes：
Judge whether the KPI Key Performance Indicator of each host is more than the adjustment threshold value respectively；
If the KPI Key Performance Indicator of the first host is more than the adjustment threshold value, using first host as host of moving out；
According to the KPI Key Performance Indicator of each host sort, select KPI Key Performance Indicator sort last specified quantity second master
Machine, as moving into host；
According to it is described move out host and it is described move into host, generate the management strategy, the management strategy by described for moving
Go out the virtual network function in host, moves to and described move into host.
5. virtual network function management method according to claim 4, which is characterized in that described according to each host
KPI Key Performance Indicator and preset load balancing, generating management strategy further includes：
Obtain the load data of the virtual deployment unit of each virtual network function in host of moving out；
According to the load data of the virtual deployment unit of each virtual network function, virtual network function is ranked up；
According to ranking results, select the virtual network function beyond the adjustment threshold value as migration target；
According to the host of moving out, the migration target and it is described move into host, generate the management strategy；The management strategy
For virtual network function that will be in the host of moving out as migration target, moves to and described move into host.
6. virtual network function management method according to claim 5, which is characterized in that before generating management strategy,
Further include：
Budget KPI Key Performance Indicator of the host after moving into the migration target is moved into described in budget；
Judge whether the budget KPI Key Performance Indicator is more than the adjustment threshold value；
If so, host is moved into gravity treatment, and correct the management strategy.
7. virtual network function management method according to claim 4, which is characterized in that after generating management strategy,
Further include：
Judge whether occur being not present host of moving out, or appearance is in the presence of host of moving out, there is no the judging results for moving into host；
If there is the judging result, the management strategy is not generated.
8. a kind of virtual network function managing device, including：Decision-making module and management module, wherein
The decision-making module is used to obtain the KPI Key Performance Indicator of each host in mainframe cluster；According to the crucial achievement of each host
Index and preset load balancing are imitated, management strategy is generated；
The management module is used to manage the virtual network function in each host according to the management strategy.
9. virtual network function managing device according to claim 8, which is characterized in that the load balancing includes
Threshold value is adjusted, the decision-making module is used to judge respectively whether the KPI Key Performance Indicator of each host to be more than the adjustment threshold value；If
The KPI Key Performance Indicator of first host is more than the adjustment threshold value, then using first host as host of moving out；According to each master
The KPI Key Performance Indicator of machine sorts, and KPI Key Performance Indicator is selected to sort the second host of last specified quantity, as moving into master
Machine；According to it is described move out host and it is described move into host, generate the management strategy, the management strategy by described for moving out
Virtual network function in host moves to and described moves into host.
10. virtual network function managing device according to claim 9, which is characterized in that the decision-making module is for obtaining
Take the load data of the virtual deployment unit of each virtual network function in moving out host；According to the dummy section of each virtual network function
The load data for affixing one's name to unit, is ranked up virtual network function；According to ranking results, void of the selection beyond the adjustment threshold value
Quasi- network function is as migration target；According to the host of moving out, the migration target and it is described move into host, generate the pipe
Reason strategy；The management strategy is used to, by the virtual network function as migration target in the host of moving out, move to described
Move into host.
11. virtual network function managing device according to claim 10, which is characterized in that the decision-making module is generating
Before management strategy, it is additionally operable to move into budget KPI Key Performance Indicator of the host after moving into the migration target described in budget；
Judge whether the budget KPI Key Performance Indicator is more than the adjustment threshold value；If so, host is moved into gravity treatment, and correct the pipe
Reason strategy.
12. according to claim 8 to 11 any one of them virtual network function managing device, which is characterized in that the decision
Between module is used for when detecting in section, the load data of each host in mainframe cluster is periodically obtained, according to each host
Load data, the KPI Key Performance Indicator of each host is calculated；Alternatively, for judging whether the dynamic equalization period arrives
It reaches, when reaching in the dynamic equalization period, the load data of each host in mainframe cluster is obtained, according to the negative of each host
Data are carried, the KPI Key Performance Indicator of each host is calculated.
13. a kind of network function virtualization system, including：Network function virtualize coordinator and virtual network function manager,
And such as claim 8 to 12 any one of them virtual network function managing device, the virtual network function managing device
For managing the corresponding virtual network function of the virtual network function manager.
14. network function virtualization system according to claim 13, which is characterized in that the virtual network function management
Device is virtualized in the network function except coordinator and virtual network function manager as autonomous device, setting；Or
The virtual network function managing device setting is inside network function virtualization coordinator；Alternatively, the virtual network
The decision-making module setting of function management device is inside network function virtualization coordinator, the virtual network function management
The management module of device is arranged inside the virtual network function manager.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710022734.3A CN108306912A (en) | 2017-01-12 | 2017-01-12 | Virtual network function management method and its device, network function virtualization system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710022734.3A CN108306912A (en) | 2017-01-12 | 2017-01-12 | Virtual network function management method and its device, network function virtualization system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108306912A true CN108306912A (en) | 2018-07-20 |
Family
ID=62872002
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710022734.3A Withdrawn CN108306912A (en) | 2017-01-12 | 2017-01-12 | Virtual network function management method and its device, network function virtualization system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108306912A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881482A (en) * | 2018-07-25 | 2018-11-23 | 新华三技术有限公司 | A kind of flow migration method, apparatus and system |
| CN109086128A (en) * | 2018-08-28 | 2018-12-25 | 迈普通信技术股份有限公司 | Method for scheduling task and device |
| CN112732409A (en) * | 2021-01-21 | 2021-04-30 | 上海交通大学 | Method and device for enabling zero-time-consumption network flow load balancing under VNF architecture |
| CN113556288A (en) * | 2021-07-23 | 2021-10-26 | 中国科学技术大学苏州高等研究院 | Device and method for updating service function link route in multi-tenant cloud |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104811396A (en) * | 2014-01-23 | 2015-07-29 | 中兴通讯股份有限公司 | Load balance (LB) method and system |
| US9450885B2 (en) * | 2012-03-26 | 2016-09-20 | Oracle International Corporation | System and method for supporting live migration of virtual machines in a virtualization environment |
| CN105978952A (en) * | 2016-04-28 | 2016-09-28 | 中国科学院计算技术研究所 | Virtualization scene flow migration method based on network function and system thereof |
-
2017
- 2017-01-12 CN CN201710022734.3A patent/CN108306912A/en not_active Withdrawn
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9450885B2 (en) * | 2012-03-26 | 2016-09-20 | Oracle International Corporation | System and method for supporting live migration of virtual machines in a virtualization environment |
| CN104811396A (en) * | 2014-01-23 | 2015-07-29 | 中兴通讯股份有限公司 | Load balance (LB) method and system |
| CN105978952A (en) * | 2016-04-28 | 2016-09-28 | 中国科学院计算技术研究所 | Virtualization scene flow migration method based on network function and system thereof |
Non-Patent Citations (2)
| Title |
|---|
| TIMOTHY WOOD等: "Sandpiper: Black-box and gray-box resource management for virtual machines", 《COMPUTER NETWORKS》 * |
| 黄冠华: "云数据中心中基于虚拟机迁移的负载均衡算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑(月刊)》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881482A (en) * | 2018-07-25 | 2018-11-23 | 新华三技术有限公司 | A kind of flow migration method, apparatus and system |
| CN108881482B (en) * | 2018-07-25 | 2021-06-29 | 新华三技术有限公司 | Traffic migration method, device and system |
| CN109086128A (en) * | 2018-08-28 | 2018-12-25 | 迈普通信技术股份有限公司 | Method for scheduling task and device |
| CN109086128B (en) * | 2018-08-28 | 2021-06-18 | 迈普通信技术股份有限公司 | Task scheduling method and device |
| CN112732409A (en) * | 2021-01-21 | 2021-04-30 | 上海交通大学 | Method and device for enabling zero-time-consumption network flow load balancing under VNF architecture |
| CN112732409B (en) * | 2021-01-21 | 2022-07-22 | 上海交通大学 | Method and device for enabling zero-time-consumption network flow load balancing under VNF architecture |
| CN113556288A (en) * | 2021-07-23 | 2021-10-26 | 中国科学技术大学苏州高等研究院 | Device and method for updating service function link route in multi-tenant cloud |
| CN113556288B (en) * | 2021-07-23 | 2022-10-28 | 中国科学技术大学苏州高等研究院 | Service function link route updating device and method in multi-tenant cloud |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3606008B1 (en) | Method and device for realizing resource scheduling | |
| Shi et al. | MDP and machine learning-based cost-optimization of dynamic resource allocation for network function virtualization | |
| CN104102543B (en) | The method and apparatus of adjustment of load in a kind of cloud computing environment | |
| CN105740074B (en) | A kind of virtual machine load-balancing method based on cloud computing | |
| US9335932B2 (en) | Storage unit selection for virtualized storage units | |
| CN102821158B (en) | A kind of method and cloud system realizing virtual machine (vm) migration | |
| CN103885831B (en) | The system of selection of virtual machine host machine and device | |
| CN110389838A (en) | A kind of Real-Time Scheduling suitable for virtual resource and online migration management-control method | |
| CN109684074A (en) | Physical machine resource allocation methods and terminal device | |
| CN105468435A (en) | NFV dynamic resource distribution method | |
| CN110515539A (en) | Cloud disk hanging method, device, equipment and storage medium based on cloud storage | |
| CN108306912A (en) | Virtual network function management method and its device, network function virtualization system | |
| CN103810047B (en) | Dynamic improves the memory affinity of logical partition | |
| CN103399781B (en) | Cloud Server and virtual machine management method thereof | |
| CN104679594B (en) | A kind of middleware distributed computing method | |
| CN104301404A (en) | Method and device for regulating resources of service system based on virtual machines | |
| CN106687926A (en) | Load balancing of cloned virtual machines | |
| CN111381928B (en) | Virtual machine migration method, cloud computing management platform and storage medium | |
| EP4029197B1 (en) | Utilizing network analytics for service provisioning | |
| CN107479947A (en) | A kind of energy consumption of virtual machine optimization method and system | |
| CN109271257A (en) | A kind of method and apparatus of virtual machine (vm) migration deployment | |
| CN103514046A (en) | Virtual machine placement method and cluster management server | |
| CN103885811B (en) | Method, system and device that dummy machine system total system is migrated online | |
| CN104539708A (en) | Capacity reduction method, device and system for cloud platform resources | |
| CN107220125A (en) | A kind of cloud resource dispatching method and device |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180720 |
|
| WW01 | Invention patent application withdrawn after publication |