CN108289113A - VBNG system multi-dummy machine load share methods and vBNG system equipments - Google Patents
VBNG system multi-dummy machine load share methods and vBNG system equipments Download PDFInfo
- Publication number
- CN108289113A CN108289113A CN201710015719.6A CN201710015719A CN108289113A CN 108289113 A CN108289113 A CN 108289113A CN 201710015719 A CN201710015719 A CN 201710015719A CN 108289113 A CN108289113 A CN 108289113A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- user
- standard grade
- capacity
- reaching
- 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- Withdrawn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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000011218 segmentation Effects 0.000 claims abstract description 50
- 239000011800 void material Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 abstract description 14
- 238000005516 engineering process Methods 0.000 description 9
- 238000007630 basic procedure Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009415 formwork Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Abstract
An embodiment of the present invention provides a kind of vBNG systems multi-dummy machine load share methods, additionally provide and carry out the vBNG system equipments that more virtual machine users share using the method.The method and equipment adjust the virtual machine of subsequent user processing according to number of users processed on each virtual machine in time, user's total capacity of each virtual machine is specifically divided into multiple segmentation capacity, user is reached the standard grade and is reached the standard grade in each virtual machine, and the user that makes as unit of being segmented capacity to reach the standard grade evenly distributes as possible between each virtual machine.Technical solution of the present invention can effectively share the number of users of the multi-dummy machine processing in multi-dummy machine vBNG systems, be adjusted in time to the virtual machine needed for subsequent user Message processing, realize the rational utilization of resource;It can be effectively reduced again because of system instability problem caused by frequent switching virtual machine.
Description
Technical field
The present invention relates to data communication technology field more particularly to virtual broadband network gateway (virtual Bro
Adband Network Gateway, vBNG) multi-dummy machine load share method and vBNG system equipments in system.
Background technology
For traditional network equipment, forwarding surface is realized with control plane on the same router, is easy restricted
In hardware CPU processing capacities, and configure underaction.For the deficiency of legacy network devices, it is proposed that utilize virtualization technology
Realize by server come handle control plane network function virtualization (Netwo rk Functions Virtualization,
NFV) technology, basic principle are the network industries that will need plurality of devices type, multiple physical equipment forms that could provide originally
Business function, is changed to externally provide business function by virtual unit, and a physical equipment may be implemented and support multiple virtual units,
Complete a variety of different kinds of business.
One important application of NFV technologies, exactly virtualizes wideband network gateway, i.e. vBNG technologies, usually using more
A virtual machine (Virtual Machine, VM) realizes the parsing to User Agreement message, and user's list item is built by virtual machine
Afterwards, then legacy network devices are handed down to, legacy network devices only need to be responsible for being forwarded flow.VBNG equipment is logical
The virtual unit obtained to the virtualization of Conventional wide band network gateway function is crossed, the hardware environment of generic server is may operate in
In, it can be achieved that the correlation function that Conventional wide band network gateway device can be realized, may include a main control unit and multiple
Retransmission unit, main control unit and retransmission unit are separately operable in different virtual machines.Terminal in user side can pass through
The retransmission unit of vBNG realizes the data interaction with network side.Multiple virtual machines can run it on a physical host simultaneously,
These virtual machines are mutually isolated and shared bottom physical resource.
Multi-dummy machine realization method would generally face the problem of sharing between multi-dummy machine, at present a solution be by
The ability of virtual machine is shared, i.e. a virtual machine at full capacity when, be just switched to another virtual machine, but this scheme
Resources of virtual machine can be caused to waste, simultaneously as single virtual machine load is excessively high to be susceptible to the problems such as unstable;Another
General scheme is directly to be shared by user, i.e., user directly shares each virtual machine, and this scheme needs to send void
Quasi- machine list item moment update, is also susceptible to unstable situation.
In Chinese patent literature CN105577723A, a kind of method for virtualizing and realizing load balancing in network is provided, is answered
It is carried out with that on the virtualization network-based control device, service traffics can be dispersed on multiple virtualization access devices,
Avoid performance bottleneck;Although above-mentioned technology relates to the problem of sharing between multi-dummy machine, but what it solved is service traffics
The Sharing of distribution is not related to user and reaches the standard grade the sharing of processing.Therefore, how multi-dummy machine load is realized in vBNG systems
Share, especially to user reach the standard grade processing realize it is effective and reasonable share, become in the NFV technologies of current data communication urgently
The technical issues of solution.
Invention content
The present invention is to solve in existing vBNG systems multi-dummy machine user reach the standard grade processing load lack effective and reasonable point
The technical issues of load, provides and realizes the method shared between virtual machine and corresponding vBNG system equipments in a kind of vBNG systems,
The method especially adapts to reach the standard grade the vBNG networks of class for user, can be timely according to number of users processed on each virtual machine
The virtual machine for adjusting subsequent user Message processing, realizes the rational utilization of resource.
A kind of vBNG systems multi-dummy machine load share method provided in an embodiment of the present invention, including:
A kind of vBNG systems multi-dummy machine load share method, the vBNG networks for class of reaching the standard grade for user, which is characterized in that
The method includes:
Receive User Agreement message;
The user capacity of reaching the standard grade of each virtual machine is configured, including the maximum number of user of each virtual machine is set separately, by single void
User's total capacity of quasi- machine is divided into several segmentation capacity, and each open ended user that reaches the standard grade of segmentation capacity of single virtual machine is arranged
Quantity summation is equal with the maximum number of user of the virtual machine, and the segmentation capacity number that each virtual machine is arranged is equal;
It configures each virtual machine and shares user and reach the standard grade, including the User Agreement message is sent on respective virtual machine and is handled,
So that the user that reaches the standard grade in single virtual machine is filled the segmentation capacity of the virtual machine one by one, keeps in each virtual machine by described
The segmentation capacity number difference that line user fills up is no more than 1.
Preferably, automatic to establish newly when the number of users of reaching the standard grade in existing each virtual machine reaches maximum number of user
Virtual machine.
As the preferred of preceding method, the maximum number of user that each virtual machine is arranged is equal, while the segmentation capacity is arranged
It is equal in magnitude.It is further preferred that the user capacity of reaching the standard grade of each virtual machine of configuration includes:
The maximum number of user of each virtual machine is set as m, the value for initializing the load sharing parameter P of each virtual machine is n,
Middle m, n are natural number, and the open ended number of users of reaching the standard grade of each segmentation capacity is m/n.
Further, it shares user as each virtual machine of the configuration and reaches the standard grade preferred, step includes:
It current loads is shared into a parameter P maximum virtual machine in flow table is set as User Agreement message and go out to connect
Mouthful, the User Agreement message is sent on the virtual machine according to the flow table and is handled, the number of users of the virtual machine reaches
It reaches the standard grade to user is stopped when (n-P+1) m/n in the virtual machine, the value of virtual machine load sharing parameter P is reduced 1;It repeats
The step is executed until the load sharing parameter P of whole virtual machines is 0 or all with having reached the standard grade per family.
As the preferred of above-mentioned selection virtual machine mode, when the load sharing parameter P of multiple virtual machines is all maximum
When value, user is reached the standard grade in the maximum virtual machine in the wherein addresses ip.
Preferably, when the load sharing parameter P of whole virtual machines is 0, new virtual machine is established automatically, by user new
It reaches the standard grade in virtual machine.
The embodiment of the present invention additionally provides a kind of vBNG system equipments, including virtual machine module, management module and load point
Module is carried on a shoulder pole, the virtual machine module includes virtual machine, wherein:
The management module is used to generate user and shares the required flow table of forwarding, and records each in the virtual machine module
Reaching the standard grade for virtual machine and is updated flow table number of users;
The management module is additionally operable to configure the user capacity of reaching the standard grade of each virtual machine, including each void is set separately
The maximum number of user of quasi- machine, is divided into several segmentation capacity by user's total capacity of single virtual machine, single virtual machine is arranged
The open ended number of users summation of reaching the standard grade of each segmentation capacity is equal with the maximum number of user of the virtual machine, and each institute is arranged
The segmentation capacity number for stating virtual machine is equal;
The management module is additionally operable to configuration user and reaches the standard grade strategy, and the user that reaches the standard grade in single virtual machine is made fill this one by one
The segmentation capacity of virtual machine, and keep the number of the segmentation capacity filled up by the user that reaches the standard grade in each virtual machine poor
No more than 1;
The load sharing module is used to be sent User Agreement message to institute according to the flow table and user strategy of reaching the standard grade
State virtual machine module.
Preferably, the number of users of reaching the standard grade that the management module is additionally operable in existing each virtual machine reaches most
When large user's number, new virtual machine is established automatically.
As the preferred of preceding solution, the maximum number of user of each virtual machine of the management module setting is homogeneous
Deng the segmentation amount of capacity is equal.It is further preferred that each virtual machine of the configuration of the management module is reached the standard grade
User capacity includes:The maximum number of user value of each virtual machine is set as m, initializes the load sharing parameter of each virtual machine
The value of P is n, and wherein m, n is natural number, and the open ended number of users of reaching the standard grade of each segmentation capacity is m/n.
Still more preferably, the configuration user of the management module reach the standard grade strategy step include:In flow table
Current loads are shared into the outgoing interface that the maximum virtual machine of parameter P is set as User Agreement message, it will according to the flow table
User reaches the standard grade in the virtual machine until the number of users of reaching the standard grade of the virtual machine reaches (n-P+1) m/n, at this time that this is virtual
The value of machine load sharing parameter P reduces 1;Above-mentioned steps are repeated until the load sharing parameter P of whole virtual machines is 0 or complete
Portion with having reached the standard grade per family.
It, will wherein addresses ip preferably, when the load sharing parameter P of multiple virtual machines is all maximum value
A maximum virtual machine is set as flow table outgoing interface, and user is reached the standard grade in the virtual machine.
Preferably, when the load sharing parameter P of whole virtual machines is 0, automatically in the virtual machine module
New virtual machine is established, user is reached the standard grade in the new virtual machine.
It realizes in the vBNG systems that previous embodiment provides and reaches the standard grade the method and apply that user shares between multi-dummy machine
The vBNG system equipments of this method can adjust the void of subsequent user processing in time according to number of users processed on each virtual machine
Quasi- machine, technical solution have the beneficial effect that:It can be effectively to the use of the multi-dummy machine processing in multi-dummy machine vBNG systems
Amount is shared, and is adjusted in time to the virtual machine needed for subsequent user Message processing, and the rationalization profit of resource is realized
With;It can be effectively reduced again because of system instability problem caused by frequent switching virtual machine.
Description of the drawings
Fig. 1 is vBNG systems multi-dummy machine load share method basic flow chart provided in an embodiment of the present invention;
Fig. 2 is the structure diagram of vBNG system equipments provided in an embodiment of the present invention;
Fig. 3 is the execution steps flow chart that vBNG system equipments provided in an embodiment of the present invention realize multi-dummy machine load sharing
Figure.
[main element symbol description]
1- routing devices;2- management modules;3- load sharing modules;4- virtual machine modules;41- virtual machines.
Specific implementation mode
To keep the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool
Body embodiment is described in detail.
The present invention provide between a kind of multi-dummy machine for class vBNG networks of reaching the standard grade suitable for user the method shared of user and
Using the vBNG system equipments of this method, subsequent user report can be adjusted in time according to number of users processed on each virtual machine
The virtual machine of text processing, realizes the rational utilization of resource.
The vBNG system multi-dummy machine load share methods of the embodiment of the present invention, the vBNG networks for class of reaching the standard grade for user,
As shown in Figure 1, its basic procedure includes:
Step S1. receives User Agreement message;
Step S2. configures the user capacity of reaching the standard grade of each virtual machine, including the maximum number of user of each virtual machine is set separately, will
User's total capacity of single virtual machine is divided into several segmentation capacity, be arranged single virtual machine each segmentation capacity it is open ended on
Line number of users summation is equal with the maximum number of user of the virtual machine, and the segmentation capacity number that each virtual machine is arranged is equal;
Each virtual machine of step S3. configurations shares user and reaches the standard grade, including the User Agreement message is sent to respective virtual machine
Upper processing, the user that reaches the standard grade in single virtual machine fill the segmentation capacity of the virtual machine, keep quilt in each virtual machine one by one
The segmentation capacity number difference that the user that reaches the standard grade fills up is no more than 1.
When the user capacity of reaching the standard grade of virtual machine is all used up, increase following steps to continue to reach the standard grade by remaining users:
Step S4. is automatic to establish newly when the number of users of reaching the standard grade in existing each virtual machine reaches maximum number of user
Virtual machine.
On the basis of above-mentioned basic procedure, further technical solution is as follows:
In above-mentioned steps S2, there are default value, initial value is the user capacity of single virtual machine in usual vBNG systems
For the true capacity of the virtual machine, user can configure the maximum number of user of single virtual machine as required, each virtual to set
User's total capacity of machine;To simplify setting and implementation procedure, and keep load sharing more uniform, the maximum of each virtual machine can be set
Number of users is equal, while it is equal that the segmentation amount of capacity is arranged.
As an embodiment, realized by the value of maximum number of user and load sharing parameter P that each virtual machine is arranged
The user capacity configuration of reaching the standard grade of multiple virtual machines.The maximum number of user value of each virtual machine is set as m;Load sharing parameter P is
Indicate the priority of each virtual machine, the parameter value is bigger, then the virtual machine shares that the priority that user reaches the standard grade is higher, i.e., user is excellent
It reaches the standard grade prior to the virtual machine.The value for initializing the load sharing parameter P of each virtual machine is n, and wherein m, n is natural number, each
The open ended number of users of reaching the standard grade of the segmentation capacity is m/n, and the integer that m is n and initial virtual machine quantity product is usually arranged
Times.
On the basis of maximum number of user and initialization load sharing parameter P is arranged, step S3 configures each virtual machine and shares
The step of user reaches the standard grade include:User is reached the standard grade in current loads share the maximum virtual machine of parameter P, as preferred
Embodiment, if the load sharing parameter P of multiple virtual machines is all maximum value, by user, the addresses ip are maximum wherein
A virtual machine in reach the standard grade;The User Agreement message is sent on the virtual machine further according to the flow table and is handled, it should
Stop user when the number of users of virtual machine reaches (n-P+1) m/n to reach the standard grade in the virtual machine, which is joined
The value of number P reduces 1;Repeat above-mentioned steps until the load sharing parameter P of whole virtual machines be 0 or all with per family on
Line;When the load sharing parameter P of whole virtual machines be 0 when, establish new virtual machine automatically, by user in new virtual machine on
Line.
It is said below by way of several embodiments to sharing the detailed process that user reaches the standard grade under different multi-dummy machine settings
It is bright.
Embodiment one
The virtual machine quantity of initial configuration is 4, the addresses ip of first to fourth virtual machine be followed successively by 190.168.1.2,
190.168.1.3,190.168.1.4 and 190.168.1.5;Each virtual machine maximum number of user of initial setting up is 1000, initially
The load sharing parameter P for changing each virtual machine is 5, starts the user that reaches the standard grade.
According to above-mentioned setting, it is equivalent to and user's total capacity of each virtual machine is divided into 5 segmentation capacity respectively, and is each
Segmentation capacity it is open ended reach the standard grade number of users be 1000/5, i.e., 200;Since the initial P values of four virtual machines are 5, user
It first reaches the standard grade from maximum 4th virtual machine of virtual machine address, starts to fill its first segmentation capacity, user reaches 200 when reaching the standard grade
When, a segmentation capacity in the 4th virtual machine all occupies, stops at this time in the 4th virtual machine and continues the user that reaches the standard grade, together
When the current P values of the 4th virtual machine are reduced to 4 by 5;Similarly, 201 to 400 user switches to that current P value is maximum and ip
The maximum third virtual machine in location is reached the standard grade, and 401 to 600 user switches to the second virtual machine and reaches the standard grade, 601 to 800 user's switching
It reaches the standard grade to the first virtual machine, so far all the P values of four virtual machines are reduced to 4, and the first of each virtual machine to cut segmentation capacity equal
Have been filled with, and so on repeat the above steps, the user that reaches the standard grade occupies second in each virtual machine segmentation capacity successively;When reaching the standard grade
When total number of users reaches 4000, the load sharing parameter P of whole virtual machines is 0, illustrates that user arrived existing virtual machine
Maximum size automatically begins to execute dilatation operation, creates new virtual machine.
Embodiment two
The virtual machine quantity of initial configuration is 2, the addresses ip of the first and second virtual machines be respectively 190.168.1.2 and
190.168.1.3;Each virtual machine maximum number of user of initial setting up is 32000, initializes the load sharing parameter P of each virtual machine
It is 2, starts the user that reaches the standard grade.
According to above-mentioned setting, it is equivalent to and user's total capacity of each virtual machine is divided into 2 segmentation capacity respectively, and is each
Segmentation capacity it is open ended reach the standard grade number of users be 32000/2, i.e., 16000;Since the initial P values of two virtual machines are 2, use
Family is first reached the standard grade from maximum second virtual machine of virtual machine address, starts to fill its first segmentation capacity, when reaching the standard grade, user reaches
When 16000, a segmentation capacity in the second virtual machine all occupies, stops at this time in the second virtual machine and continues use of reaching the standard grade
Family, while the current P values of the second virtual machine are reduced to 1 by 2;Similarly, 16001 to 32000 user is on the first virtual machine
After line, the P values of two virtual machines are 1, and so on are repeated the above steps, the user that reaches the standard grade occupies in each virtual machine successively
Two segmentation capacity;And so on, when total number of users of reaching the standard grade reaches 64000, the load sharing parameter P of two virtual machines subtracts
It is 0, illustrates that user arrived the maximum size of existing virtual machine, automatically begins to execute dilatation operation, create new virtual machine.
Embodiment three
When the virtual machine quantity of initial configuration is 1, the initial setting up virtual machine maximum number of user is 2000, and initialization is negative
It is 1 that lotus, which shares parameter P, starts the user that reaches the standard grade;Since only there are one virtual machine, the limiting case of as single virtual machine equally may be used
Operation is executed according to the flow of embodiment one, two, user directly reaches the standard grade in the virtual machine, without switching over, when number of users reaches
When to 2000, the user that reaches the standard grade arrived the maximum size of single virtual machine, automatically begins to execute dilatation operation, create new virtual
Machine.
Those of ordinary skill in the art it would be appreciated that, realize that all or part of step of above method embodiment can lead to
The relevant hardware of program instruction is crossed to complete, foregoing routine can be stored in one as the computer of server and storage can be read
In medium.
VBNG system equipments provided in an embodiment of the present invention apply above-mentioned vBNG systems multi-dummy machine load sharing side
Method, including operation have virtual machine module 4, management module 2 and the load sharing module 3 of virtual machine 41.
In the present embodiment, as shown in Figure 2, it is shown that connecting substantially when the vBNG system equipments are connect with routing device 1
Relationship is connect, including:
Load sharing module 3 is connect with routing device 1, management module 2 and virtual machine module 4;Load sharing module 3 receives
User Agreement message from routing device 1, and under the control of management module 2, User Agreement message is issued to virtual machine module 4
The virtual machine 41 for meeting condition is handled;Specifically:
Management module 2 is used to generate user and shares the required flow table of forwarding, records the number of users of reaching the standard grade of each virtual machine, and right
Flow table is updated;It is additionally operable to configure the user capacity of reaching the standard grade of each virtual machine, includes the maximum number of user of each virtual machine of setting, it will
User's total capacity of each virtual machine is divided into the segmentation capacity of same number, and same virtual machine is respectively segmented capacity summation and is used with maximum
Amount is equal;It is additionally operable to configuration user to reach the standard grade strategy, so that user is reached the standard grade in the virtual machine of virtual machine module 4 and is occupied one by one and divide
Duan Rongliang keeps each virtual machine to be no more than 1 by the number difference for the segmentation capacity that user occupies completely;
The flow table and user that load sharing module 3 is used to be provided according to management module 2 are reached the standard grade tactful by User Agreement message
It sends to the respective virtual machine of virtual machine module 4 and is handled.
Management module 2 is additionally operable to when the number of users of reaching the standard grade in existing each virtual machine reaches maximum number of user, from
It is dynamic to establish new virtual machine.
With reference to the embodiment of preceding method, management module 2 can set each virtual machine with equal maximum number of user and
It is segmented capacity;The concrete mode that management module 2 configures the user capacity of reaching the standard grade of multiple virtual machines is:The maximum of each virtual machine is used
Amount value is set as m, and the value for initializing the load sharing parameter P of each virtual machine is n, and wherein m, n is natural number, each described
The open ended number of users of reaching the standard grade of segmentation capacity is m/n;Management module 2 configure user reach the standard grade strategy concrete mode be:It will be current
It shares the maximum virtual machine of parameter P and is set as flow table outgoing interface, user is reached the standard grade in the virtual machine up to the virtual machine
Until number of users of reaching the standard grade reaches (n-P+1) m/n, the value which is shared to parameter P at this time reduces 1;Repeat this
Step is until the parameter P that shares of whole virtual machines is 0 or all with having reached the standard grade per family;Parameter is shared described in multiple virtual machines
When P is all maximum value, the wherein maximum virtual machine in the addresses ip is set to flow table outgoing interface, by user on the virtual machine
Line;When whole virtual machines share parameter P be 0 when, new virtual machine is established in the virtual machine module 4 automatically, by user
It reaches the standard grade in the new virtual machine.
Realize that the flow of multi-dummy machine load sharing illustrates with vBNG system equipments provided in an embodiment of the present invention:Make
Reached the standard grade with conventional router and virtual unit configuration user, be set as needed parameter, including each virtual machine user capacity (most
Large user's number) i.e. and initial load sharing parameter P, after setting parameter, start the user that reaches the standard grade;Use between each virtual machine
Family is shared to carry out according to step shown in Fig. 3:
(1) maximum number of user that management module configures each virtual machine is m;
(2) each virtual machine load sharing parameter P is initialized as n by management module;
(3) management module compares load sharing parameter P, selects the maximum virtual machine of P values that flow table is written, goes out to connect as flow table
Mouthful;If P values are identical, compare the addresses ip, flow table is written in the virtual machine for selecting the addresses ip big, as flow table outgoing interface;
(4) number of users of reaching the standard grade of management module records current virtual machine, and the number of users is compared with standard value, it marks
Quasi- value initial value is m/n;If being less than the standard value, continuation is reached the standard grade on this virtual machine;Otherwise, by the load of the virtual machine
The value for sharing parameter P subtracts 1, repeats aforementioned 3rd step, and selection switches to other virtual machines and continues the user that reaches the standard grade;
(5) when the number of users of reaching the standard grade of all virtual machines all reaches m/n, the second wheel is proceeded by from aforementioned 3rd step
User reaches the standard grade, and the standard value in the 4th step is increased m/n, i.e., is compared virtual machine number of users of reaching the standard grade with 2m/n;Class according to this
It pushes away, in often taking turns, user's standard value of reaching the standard grade increases m/n, and the load sharing parameter P values of each virtual machine subtract 1, until load sharing is joined
Number becomes 0;
(6) when the load sharing parameter P of all virtual machines is 0, then automatic dilatation operation is carried out to virtual machine, established
New virtual machine.
Above-mentioned vBNG system equipments can adjust subsequent user processing in time according to number of users processed on each virtual machine
Virtual machine, realize the rational utilization of resource;Routing device usually can be used traditional router, and the management mould
Block, load sharing module and virtual machine module can use virtualization technology, with void on the server as hardware carrier
The form work of quasi- machine.
Above-mentioned vBNG system equipments embodiment is schematical, and the common sense such as well known structure and characteristic do not describe excessively;
Each embodiment is described by the way of progressive, and just to refer each other for identical similar portion between each embodiment, institute in each embodiment
The technical characteristic being related to can be combined with each other under the premise of not constituting conflict between each other;Wherein discrete parts is used as to carry out
The unit of explanation can may not be to be physically separated, and the component presented as a unit can be on physical layer
Positioned at the same place, can also be distributed in multiple networks;Part therein or complete can be selected according to the actual needs
Portion module achieves the purpose of the solution of this embodiment.Those of ordinary skill in the art without creative efforts,
It can understand and implement technical solution of the present invention.
In the description of the above embodiment of the present invention, term " first ", " second " etc. are used for description purposes only, without that should manage
Solution is instruction or implies relative importance.
It should be pointed out that for those skilled in the art, before not departing from principle of the present invention
It puts, can also make several improvements and retouch, these improvements and modifications also should be regarded as falling into protection scope of the present invention.
Claims (14)
1. a kind of vBNG systems multi-dummy machine load share method, the vBNG networks for class of reaching the standard grade for user, which is characterized in that institute
The method of stating includes:
Receive User Agreement message;
The user capacity of reaching the standard grade of each virtual machine is configured, including the maximum number of user of each virtual machine is set separately, by single virtual machine
User's total capacity be divided into several segmentation capacity, each open ended number of users of reaching the standard grade of segmentation capacity of single virtual machine is set
Summation is equal with the maximum number of user of the virtual machine, and the segmentation capacity number that each virtual machine is arranged is equal;
It configures each virtual machine and shares user and reach the standard grade, including the User Agreement message is sent on respective virtual machine and is handled, make list
The user that reaches the standard grade in one virtual machine fills the segmentation capacity of the virtual machine one by one, keeps in each virtual machine by the use of reaching the standard grade
The segmentation capacity number difference that family is filled up is no more than 1.
2. according to the method described in claim 1, it is characterized in that, the method further includes:
When the number of users of reaching the standard grade in existing each virtual machine reaches maximum number of user, new virtual machine is established automatically.
3. method according to claim 1 or 2, which is characterized in that the maximum number of user of setting each virtual machine is homogeneous
Deng the segmentation amount of capacity is equal.
4. according to the method described in claim 3, it is characterized in that, each virtual machine of the configuration reach the standard grade user capacity the step of
Including:
The maximum number of user of each virtual machine is set as m, initialize the load sharing parameter P of each virtual machine value be n, wherein m,
N is natural number, and the open ended number of users of reaching the standard grade of each segmentation capacity is m/n.
5. according to the method described in claim 4, it is characterized in that, each virtual machine of configuration is shared the step of user reaches the standard grade and is wrapped
It includes:
Current loads are shared into the outgoing interface that the maximum virtual machine of parameter P is set as User Agreement message, root in flow table
The User Agreement message is sent on the virtual machine according to the flow table and is handled, the number of users of the virtual machine reaches (n-P
+ 1) stop user when m/n to reach the standard grade in the virtual machine, the value of virtual machine load sharing parameter P is reduced 1;Repeat this
Step is until the load sharing parameter P of whole virtual machines is 0 or all with having reached the standard grade per family.
6. according to the method described in claim 5, it is characterized in that, the load sharing parameter P when multiple virtual machines is all
When maximum value, user is reached the standard grade in the maximum virtual machine in the wherein addresses ip.
7. according to the method described in claim 5, it is characterized in that, further including following steps:
When the load sharing parameter P of whole virtual machines be 0 when, establish new virtual machine automatically, by user in new virtual machine on
Line.
8. a kind of vBNG system equipments, including virtual machine module, the virtual machine module include virtual machine, which is characterized in that institute
It further includes management module and load sharing module to state equipment, wherein:
The management module is used to generate user and shares the required flow table of forwarding, and records each void in the virtual machine module
Reaching the standard grade for quasi- machine and is updated flow table number of users;
The management module is additionally operable to configure the user capacity of reaching the standard grade of each virtual machine, including each virtual machine is set separately
Maximum number of user, user's total capacity of single virtual machine is divided into several segmentation capacity, each institute of single virtual machine is set
It is equal with the maximum number of user of the virtual machine to state the segmentation open ended number of users summation of reaching the standard grade of capacity, each void is set
The segmentation capacity number of quasi- machine is equal;
The management module is additionally operable to configuration user and reaches the standard grade strategy, so that the user that reaches the standard grade in single virtual machine is filled this one by one virtual
The segmentation capacity of machine, and keep not surpassed by the number difference for the segmentation capacity that the user that reaches the standard grade fills up in each virtual machine
Cross 1;
The load sharing module is used to be sent User Agreement message to the void according to the flow table and user strategy of reaching the standard grade
Quasi- machine module.
9. equipment according to claim 8, which is characterized in that the management module is additionally operable to existing each described virtual
When number of users of reaching the standard grade in machine reaches maximum number of user, new virtual machine is established automatically.
10. equipment according to claim 8 or claim 9, which is characterized in that each virtual machine of the management module setting
Maximum number of user is equal, and the segmentation amount of capacity is equal.
11. equipment according to claim 10, which is characterized in that each virtual machine of the configuration of the management module
User capacity of reaching the standard grade include:The maximum number of user value of each virtual machine is set as m, initializes the load point of each virtual machine
The value for carrying on a shoulder pole parameter P is n, and wherein m, n is natural number, and the open ended number of users of reaching the standard grade of each segmentation capacity is m/n.
12. equipment according to claim 11, which is characterized in that the configuration user of the management module reaches the standard grade strategy
The step of include:Current loads are shared into the maximum virtual machine of parameter P in flow table and are set as going out for User Agreement message
Interface reaches the standard grade user until the number of users of reaching the standard grade of the virtual machine reaches (n-P+1) in the virtual machine according to the flow table
Until m/n, the value of virtual machine load sharing parameter P is reduced 1 at this time;Above-mentioned steps are repeated up to whole virtual machines
Load sharing parameter P is 0 or all with having reached the standard grade per family.
13. equipment according to claim 12, which is characterized in that the configuration user of the management module reaches the standard grade strategy
Further include:It, will wherein maximum one of addresses ip when the load sharing parameter P of multiple virtual machines is all maximum value
Virtual machine is set as flow table outgoing interface, and user is reached the standard grade in the virtual machine.
14. equipment according to claim 12, which is characterized in that the configuration user of the management module reaches the standard grade strategy
Further include:When the load sharing parameter P of whole virtual machines is 0, new void is established in the virtual machine module automatically
Quasi- machine, user is reached the standard grade in the new virtual machine.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710015719.6A CN108289113A (en) | 2017-01-10 | 2017-01-10 | VBNG system multi-dummy machine load share methods and vBNG system equipments |
PCT/CN2017/103039 WO2018129957A1 (en) | 2017-01-10 | 2017-09-22 | Vbng system multi-virtual machine load sharing method and vbng system device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710015719.6A CN108289113A (en) | 2017-01-10 | 2017-01-10 | VBNG system multi-dummy machine load share methods and vBNG system equipments |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108289113A true CN108289113A (en) | 2018-07-17 |
Family
ID=62819902
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710015719.6A Withdrawn CN108289113A (en) | 2017-01-10 | 2017-01-10 | VBNG system multi-dummy machine load share methods and vBNG system equipments |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108289113A (en) |
WO (1) | WO2018129957A1 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10560331B2 (en) | 2018-02-07 | 2020-02-11 | Juniper Networks, Inc. | Self-driven and adaptable multi-vBNG management orchestration |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2629490A1 (en) * | 2012-02-14 | 2013-08-21 | Telefonaktiebolaget L M Ericsson (Publ) | Optimizing traffic load in a communications network |
CN103812788A (en) * | 2012-11-05 | 2014-05-21 | 华为技术有限公司 | Method, device and system for controlling user to access |
CN103902353A (en) * | 2014-03-31 | 2014-07-02 | 华为技术有限公司 | Virtual machine deployment method and device |
CN103984602A (en) * | 2014-05-20 | 2014-08-13 | 华为技术有限公司 | VM (virtual machine) resource scheduling method, device and system |
CN104796469A (en) * | 2015-04-15 | 2015-07-22 | 北京中油瑞飞信息技术有限责任公司 | Method and device for configuring cloud computing platforms |
EP2988214A1 (en) * | 2014-08-20 | 2016-02-24 | Alcatel Lucent | Method for balancing a load, a system, an elasticity manager and a computer program product |
CN106254447A (en) * | 2016-07-29 | 2016-12-21 | 北京华为数字技术有限公司 | Load-balancing method, BRAS equipment and SDN controller |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8719069B2 (en) * | 2009-07-23 | 2014-05-06 | Brocade Communications Systems, Inc. | Method and apparatus for providing virtual machine information to a network interface |
CN103297352B (en) * | 2012-02-23 | 2017-03-15 | 华为技术有限公司 | A kind of collocation method of access strategy and wideband network gateway |
CN104579732B (en) * | 2013-10-21 | 2018-06-26 | 华为技术有限公司 | Virtualize management method, the device and system of network function network element |
CN104702521A (en) * | 2013-12-06 | 2015-06-10 | 中兴通讯股份有限公司 | Load balance method and device |
-
2017
- 2017-01-10 CN CN201710015719.6A patent/CN108289113A/en not_active Withdrawn
- 2017-09-22 WO PCT/CN2017/103039 patent/WO2018129957A1/en active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2629490A1 (en) * | 2012-02-14 | 2013-08-21 | Telefonaktiebolaget L M Ericsson (Publ) | Optimizing traffic load in a communications network |
CN103812788A (en) * | 2012-11-05 | 2014-05-21 | 华为技术有限公司 | Method, device and system for controlling user to access |
CN103902353A (en) * | 2014-03-31 | 2014-07-02 | 华为技术有限公司 | Virtual machine deployment method and device |
CN103984602A (en) * | 2014-05-20 | 2014-08-13 | 华为技术有限公司 | VM (virtual machine) resource scheduling method, device and system |
EP2988214A1 (en) * | 2014-08-20 | 2016-02-24 | Alcatel Lucent | Method for balancing a load, a system, an elasticity manager and a computer program product |
CN104796469A (en) * | 2015-04-15 | 2015-07-22 | 北京中油瑞飞信息技术有限责任公司 | Method and device for configuring cloud computing platforms |
CN106254447A (en) * | 2016-07-29 | 2016-12-21 | 北京华为数字技术有限公司 | Load-balancing method, BRAS equipment and SDN controller |
Non-Patent Citations (1)
Title |
---|
通信学报: "考虑虚拟机间性能互扰的虚拟资源分配方法", 《通信学报》 * |
Also Published As
Publication number | Publication date |
---|---|
WO2018129957A1 (en) | 2018-07-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115380514B (en) | Automatic deployment of network elements for heterogeneous computing elements | |
US11625154B2 (en) | Stage upgrade of image versions on devices in a cluster | |
US11863352B2 (en) | Hierarchical networking for nested container clusters | |
US11502958B2 (en) | Automatic configuration of logical routers on edge nodes | |
US9887959B2 (en) | Methods and system for allocating an IP address for an instance in a network function virtualization (NFV) system | |
US9274851B2 (en) | Core-trunking across cores on physically separated processors allocated to a virtual machine based on configuration information including context information for virtual machines | |
US8483087B2 (en) | Port pooling | |
US7751409B1 (en) | Logical service domains for enabling network mobility | |
CN102118319B (en) | Traffic load balancing method and device | |
CN110662231A (en) | Network slice resource adjusting method and system for 5G environment | |
JP2018523932A (en) | Load balancing computing devices, systems, and methods | |
KR20170070143A (en) | Software-defined network-based method and system for implementing content distribution network | |
CN104521198A (en) | System and method for virtual ethernet interface binding | |
CN108111383A (en) | A kind of cross-domain container virtual network construction method based on SDN | |
US20140122743A1 (en) | Shared interface among multiple compute units | |
CN104852859B (en) | A kind of aggregation interface method for processing business and equipment | |
US20230028837A1 (en) | Scaling for split-networking datapath | |
CN104539558A (en) | Capacity-expansible IP telephone exchange blade mechanism frame and automatic capacity expansion method | |
CN110830574A (en) | Method for realizing intranet load balance based on docker container | |
CN109525413B (en) | CDN network function virtualization management method, device and system | |
US11838206B2 (en) | Edge node with datapath split between pods | |
US20190042294A1 (en) | System and method for implementing virtualized network functions with a shared memory pool | |
CN108289113A (en) | VBNG system multi-dummy machine load share methods and vBNG system equipments | |
CN106656717B (en) | A method of dividing network domains | |
Rodriguez et al. | Live migration in green virtualized networks |
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: 20180717 |
|
WW01 | Invention patent application withdrawn after publication |