WO2018129957A1 - Vbng system multi-virtual machine load sharing method and vbng system device - Google Patents

Abstract

Embodiments of the present invention provide a vBNG system multi-virtual machine load sharing method, and also provide a vBNG system device for multi-virtual machine user sharing by using the method. The method and device adjust virtual machines subsequently processed by users in time according to the number of users processed on each virtual machine, segment the user total capacity of each virtual machine into a plurality of segmented capacities, enable users to be online in each virtual machine, and evenly distribute the online users between the virtual machines in segmented capacities as far as possible. The technical solution of the present invention can effectively share the number of users processing multiple virtual machines in the multi-virtual machine vBNG system, adjust the virtual machines required for subsequent user message processing in time to achieve rational utilization of resources, and can effectively reduce the system unstable problem caused by frequently switching the virtual machines.

Description

vBNG system multi-virtual machine load sharing method and vBNG system equipment Technical field

The present disclosure relates to the field of data communication technologies, and in particular, to a multi-virtual machine load sharing method and a vBNG system device in a virtual broadband network gateway (vBNG) system.

Background technique

For the traditional network equipment, the forwarding plane and the control plane are implemented on the same router, which is easily subject to the hardware CPU processing capability and the configuration is not flexible enough. Aiming at the shortcomings of traditional network equipment, this paper proposes the use of virtualization technology to realize the network function virtualization (NFV) technology that the server handles the control plane. The basic principle is that it will require multiple device types and multiple physical devices. The network service function provided by the form can be used to provide service functions to the virtual device. The physical device can support multiple virtual devices and complete multiple types of services.

An important application of the NFV technology is to virtualize the broadband network gateway, that is, the vBNG technology, which usually uses multiple virtual machines (VMs) to parse the user protocol packets. After the virtual machine constructs the user entries, It is then sent to traditional network devices. Traditional network devices only need to be responsible for forwarding traffic. The vBNG device is a virtual device obtained by virtualizing the function of the traditional broadband network gateway, and can be run in a hardware environment of a general-purpose server, and can implement related functions that can be implemented by the traditional broadband network gateway device, and can include a main control unit and multiple The forwarding unit, the main control unit and the forwarding unit respectively run in different virtual machines. The terminal on the user side can implement data interaction with the network side through the forwarding unit of the vBNG. Multiple virtual machines can run on a single physical host simultaneously, which are isolated from each other and share the underlying physical resources.

Multi-virtual machine implementations often face the problem of sharing between multiple virtual machines. Currently, one solution is to share the capacity of the virtual machine, that is, when one virtual machine is fully loaded, it switches to another virtual machine, but This kind of solution will cause waste of virtual machine resources, and at the same time, because of the high load of a single virtual machine, it is easy to be unstable. Another common solution is to share the user directly, that is, the user directly shares the virtual machine. The virtual machine entries need to be updated at any time, and it is prone to instability.

In the Chinese patent document CN105577723A, a method for implementing load sharing in a virtualized network is provided, which is applied to a controller of the virtualized network, and the service traffic distribution can be distributed on multiple virtualized access devices, thereby avoiding Performance bottlenecks; the above-mentioned technologies involve the sharing of multiple virtual machines, but they solve the problem of sharing traffic distribution, and do not involve the sharing of user online processing. Therefore, how to realize the sharing of multiple virtual machine loads in the vBNG system, especially the effective and reasonable sharing of user online processing, has become an urgent technical problem in the current data communication NFV technology.

Summary of the invention

The disclosure of the present disclosure is to solve the problem that the multi-virtual machine user online processing load is insufficiently and reasonably shared in the related vBNG system. The technical problem is to provide a method for implementing virtual machine sharing in a vBNG system and a corresponding vBNG system device. The method is particularly suitable for a vBNG network for a user to go online, and can be based on the number of users processed on each virtual machine. The virtual machine for subsequent user message processing is adjusted in time to realize rational utilization of resources.

A multi-virtual machine load sharing method for a vBNG system provided by an embodiment of the present disclosure includes:

A vBNG system multi-virtual machine load sharing method is used for a vBNG network of a user online class, wherein the method includes:

Receiving user protocol packets;

Configure the online user capacity of each virtual machine, including setting the maximum number of users for each virtual machine, dividing the total user capacity of a single virtual machine into a number of segment sizes, and setting the online users that can be accommodated by each segment capacity of a single virtual machine. The sum of the numbers is equal to the maximum number of users of the virtual machine, and the number of segmented capacities of each virtual machine is set to be equal;

Configuring each virtual machine to share the user's online connection, including sending the user protocol packet to the corresponding virtual machine for processing, so that the online user in the single virtual machine fills the segment capacity of the virtual machine one by one, and keeps the virtual machine in the virtual machine. The difference in the number of segmented capacities filled by the online user does not exceed one.

Preferably, when the number of online users in each existing virtual machine reaches the maximum number of users, a new virtual machine is automatically established.

As a preferred method of the foregoing method, the maximum number of users of each virtual machine is set equal, and the segment size is set to be equal. In an embodiment, the online user capacity for configuring each virtual machine includes:

The maximum number of users of each virtual machine is set to m, and the value of the load sharing parameter P of each virtual machine is initialized, where m and n are natural numbers, and the number of online users that can be accommodated in each of the segmented capacities is m. /n.

In an embodiment, as a configuration, each virtual machine shares the preference of the user to go online, and the steps include:

Setting a virtual machine whose current load sharing parameter P is the largest in the flow table as an outgoing interface of the user protocol packet, and sending the user protocol packet to the virtual machine according to the flow table for processing, the virtual machine When the number of users reaches (n-P+1)·m/n, the user is stopped on the virtual machine, and the value of the virtual machine load sharing parameter P is decreased by one. This step is repeated until the load sharing parameter P of all virtual machines is repeated. 0 or all users are online.

Preferably, when the load sharing parameter P of the plurality of virtual machines is the maximum value, the user is brought online in a virtual machine in which the ip address is the largest.

Preferably, when the load sharing parameter P of all the virtual machines is 0, a new virtual machine is automatically established, and the user is brought online in the new virtual machine.

The embodiment of the present disclosure further provides a vBNG system device, including a virtual machine module, a management module, and a load sharing module, where the virtual machine module includes a virtual machine, where:

The management module is configured to generate a flow table required for the user to share the forwarding, and record the number of online users of each virtual machine in the virtual machine module and update the flow table;

The management module is further configured to configure an online user capacity of each of the virtual machines, including setting a maximum number of users of each virtual machine, dividing a total user capacity of the single virtual machine into a plurality of segment capacities, and setting a single virtual The sum of the number of online users that can be accommodated by each of the segment capacities of the machine is equal to the maximum number of users of the virtual machine, and each of the settings is set. The number of segmentation capacities of the virtual machine is equal;

The management module is further configured to configure a user online policy, so that the online users in the single virtual machine fill the segment capacity of the virtual machine one by one, and keep the segments of each virtual machine filled by the online users. The difference in the number of capacities does not exceed 1;

The load sharing module is configured to distribute user protocol packets to the virtual machine module according to the flow table and the user online policy.

Preferably, the management module is further configured to automatically establish a new virtual machine when the number of online users in each of the existing virtual machines reaches a maximum number of users.

Preferably, as described in the foregoing technical solution, the maximum number of users of each virtual machine set by the management module is equal, and the segment size is equal in size. In an embodiment, preferably, the configuring the online user capacity of each of the virtual machines includes: setting a maximum user value of each virtual machine to m, and initializing load sharing of each virtual machine. The value of the parameter P is n, where m and n are both natural numbers, and the number of online users that each of the segment sizes can accommodate is m/n.

In an embodiment, the step of configuring the user online policy of the management module includes: setting, in the flow table, a virtual machine whose current load sharing parameter P is the largest as an outbound interface of the user protocol packet, according to The flow table is used by the user to go online until the number of online users of the virtual machine reaches (n-P+1)·m/n. At this time, the value of the virtual machine load sharing parameter P is decreased by one; Perform the above steps until the load sharing parameter P of all virtual machines is 0 or all users are online.

Preferably, when the load sharing parameter P of the plurality of virtual machines is the maximum value, one virtual machine with the largest IP address is set as the flow table out interface, and the user is online on the virtual machine.

Preferably, when the load sharing parameter P of all the virtual machines is 0, a new virtual machine is automatically established in the virtual machine module, and the user is brought online in the new virtual machine.

The embodiment of the present disclosure further provides a storage medium, configured to store a program code, where the program code is configured to perform the vBNG system multi-virtual machine load sharing method according to any one of the above.

The method for implementing on-line user sharing between multiple virtual machines in the vBNG system provided by the foregoing embodiment, and the vBNG system device applying the method, can timely adjust the virtual machine processed by the subsequent user according to the number of processed users on each virtual machine. The technical solution has the beneficial effects that the number of users processed by multiple virtual machines in the virtual machine vBNG system can be effectively shared, and the virtual machines required for subsequent user message processing are adjusted in time to realize rational utilization of resources. It can effectively reduce the system instability caused by frequent switching of virtual machines.

DRAWINGS

FIG. 1 is a basic flowchart of a multi-virtual machine load sharing method for a vBNG system according to an embodiment of the present disclosure;

2 is a structural block diagram of a vBNG system device according to an embodiment of the present disclosure;

FIG. 3 is a flowchart of an execution step of implementing multi-virtual machine load sharing by a vBNG system device according to an embodiment of the present disclosure.

The main component symbol description:

1-Routing device; 2-Management module; 3-Load sharing module; 4-Virtual machine module; 41-Virtual machine.

detailed description

The technical problems, the technical solutions, and the advantages of the present invention will be more clearly described in conjunction with the accompanying drawings and exemplary embodiments.

The present disclosure provides a method for user sharing between multiple virtual machines in a user-type vBNG network, and a vBNG system device that applies the method, and can timely adjust subsequent user packet processing according to the number of processed users on each virtual machine. Virtual machines to achieve rational use of resources.

The vBNG system multi-virtual machine load sharing method of the embodiment of the present disclosure is used for a vBNG network of a user online class. As shown in FIG. 1 , the basic process includes:

Step S1, receiving a user protocol message;

Step S2, configuring the online user capacity of each virtual machine, including setting the maximum number of users of each virtual machine separately, dividing the total user capacity of the single virtual machine into a plurality of segmented capacities, and setting each segment capacity of the single virtual machine to accommodate The sum of the number of online users is equal to the maximum number of users of the virtual machine, and the number of segmented capacities of each virtual machine is set to be equal;

Step S3, configuring each virtual machine to share the user online, including sending the user protocol packet to the corresponding virtual machine for processing, and the online user in the single virtual machine fills the segment capacity of the virtual machine one by one, and maintains each virtual machine. The difference in the number of segment sizes filled by the online users is not more than one.

When the online user capacity of the virtual machine is completely exhausted, add the following steps to continue the online users:

In step S4, when the number of online users in each existing virtual machine reaches the maximum number of users, a new virtual machine is automatically established.

Based on the above basic process, in an embodiment, the technical solution is as follows:

In the above step S2, the user capacity of a single virtual machine in the vBNG system has a default value, and the initial value is the real capacity of the virtual machine. The user can configure the maximum number of users of a single virtual machine as needed, thereby setting each virtual machine. In order to simplify the setting and execution process and make the load sharing more uniform, the maximum number of users of each virtual machine can be set equal, and the segmented capacity is set to be equal.

As an embodiment, the online user capacity configuration of the plurality of virtual machines is implemented by setting the maximum number of users of each virtual machine and the value of the load sharing parameter P. The maximum user value of each virtual machine is set to m. The load sharing parameter P indicates the priority of each virtual machine. If the value of the parameter is larger, the priority of the virtual machine sharing user online is higher, that is, the user takes precedence over the virtual machine. The virtual machine is online. The value of the load sharing parameter P for initializing each virtual machine is n, where m and n are both natural numbers, and the number of online users that can be accommodated in each of the segment sizes is m/n, and m is usually set to n and the initial virtual machine. An integer multiple of the quantity product.

The step of setting the maximum number of users and initializing the load sharing parameter P, the step of configuring the virtual machine to share the user online in step S3 includes: putting the user online in a virtual machine with the largest load sharing parameter P as the preferred embodiment, If the load sharing parameter P of the plurality of virtual machines is the maximum value, the user is in the ip The user is sent to the virtual machine for processing according to the flow table, and the number of users of the virtual machine reaches (n-P+1)·m/n. Stop the user from going online on the virtual machine and reduce the value of the virtual machine load sharing parameter P by one. Repeat the above steps until the load sharing parameter P of all virtual machines is 0 or all users are online; when all virtual machines are loaded When the sharing parameter P is 0, a new virtual machine is automatically created, and the user is brought online in the new virtual machine.

The following describes an exemplary flow of sharing user online in different multi-virtual machine settings through several embodiments.

Embodiment 1

The initial number of virtual machines is 4, and the IP addresses of the first to fourth virtual machines are 190.168.1.2, 190.168.1.3, 190.168.1.4, and 190.168.1.5. The maximum number of users for each virtual machine is 1000, which is initialized. The load sharing parameter P of each virtual machine is 5, and the online user is started.

According to the above setting, the total user capacity of each virtual machine is divided into five segment capacities, and the number of online users that can be accommodated in each segment capacity is 1000/5, that is, 200; since four virtual machines are initially The P value is 5, the user starts to fill the first segment capacity from the fourth virtual machine with the largest virtual machine address, and when the online user reaches 200, the segment capacity of the fourth virtual machine is all Occupied, stop the online user in the fourth virtual machine, and reduce the current P value of the fourth virtual machine from 5 to 4. Similarly, the users 201 to 400 switch to the current P value and the largest IP address. The third virtual machine is online, the users of 401 to 600 switch to the second virtual machine, and the users of 601 to 800 switch to the first virtual machine, and the P values of all four virtual machines are reduced to four, and the virtual machines of each virtual machine are reduced to four. The first segmentation segment capacity is full, and the above steps are repeated. The online users occupy the second segment capacity in each virtual machine in turn; when the total number of online users reaches 4000, the load sharing parameters of all virtual machines are P. 0, explain Has reached the maximum capacity of existing virtual machines, automatic expansion started operations, creating a new virtual machine.

Embodiment 2

The number of virtual machines that are initially configured is 2. The IP addresses of the first and second virtual machines are 190.168.1.2 and 190.168.1.3 respectively. The maximum number of users for each virtual machine is 32000. The load sharing parameters of each virtual machine are initialized. For 2, start online users.

According to the above setting, the total user capacity of each virtual machine is divided into two segment capacities, and the number of online users that can be accommodated in each segment capacity is 32000/2, that is, 16000; The P value is 2, the user starts to fill the first segment capacity from the second virtual machine with the largest virtual machine address, and when the online user reaches 16000, the segment capacity of the second virtual machine is all Occupied, stop the online user in the second virtual machine, and reduce the current P value of the second virtual machine from 2 to 1. Similarly, the users of 1600-1 to 32000 are online after the first virtual machine is online. The P value of the machine is all 1, and the above steps are repeated, and the online users occupy the second segment capacity in each virtual machine in turn; and so on, when the total number of online users reaches 64000, the load sharing of the two virtual machines The parameter P is reduced to 0, indicating that the user has reached the upper limit of the capacity of the existing virtual machine, and automatically starts the expansion operation to create a new virtual machine.

Embodiment 3

When the number of virtual machines that are initially configured is 1, the maximum number of users for the virtual machine is initially set to 2000, and the initialization is negative. The load-bearing parameter P is 1, and the online user is started. Since there is only one virtual machine, which is the limit of the single virtual machine, the operation can be performed according to the processes of the first and second embodiments, and the user directly goes online on the virtual machine without switching. When the number of users reaches 2000, the online user has reached the upper limit of the capacity of a single virtual machine, and automatically starts the expansion operation to create a new virtual machine.

It should be understood by those skilled in the art that all or part of the steps of implementing the foregoing method embodiments may be performed by hardware related to program instructions, which may be stored in a readable storage medium of a computer as a server.

The vBNG system device provided by the embodiment of the present disclosure applies the vBNG system multi-virtual machine load sharing method, including the virtual machine module 4, the management module 2, and the load sharing module 3 running the virtual machine 41.

In this embodiment, as shown in FIG. 2, the basic connection relationship between the vBNG system device and the routing device 1 is displayed, including:

The load sharing module 3 is connected to the routing device 1, the management module 2, and the virtual machine module 4. The load sharing module 3 receives the user protocol packet from the routing device 1, and delivers the user protocol packet to the virtual network under the control of the management module 2. The machine module 4 satisfies the conditional virtual machine 41 for processing; exemplary:

The management module 2 is configured to generate a flow table required for the user to share the forwarding, record the number of online users of each virtual machine, and update the flow table; and set the online user capacity of each virtual machine, including setting the maximum size of each virtual machine. The number of users is divided into the same number of segmented capacity of each virtual machine. The sum of the capacity of each segment of the same virtual machine is equal to the maximum number of users. It is also configured to configure the user online policy to make the user in the virtual machine module 4 The virtual machine is online and occupies the segmentation capacity one by one, and the difference in the number of segmentation capacities that each virtual machine is completely occupied by the user does not exceed one;

The load sharing module 3 is configured to distribute the user protocol packets to the corresponding virtual machines of the virtual machine module 4 according to the flow table provided by the management module 2 and the user online policy.

The management module 2 is further configured to automatically create a new virtual machine when the number of online users in each existing virtual machine reaches the maximum number of users.

Referring to the implementation manner of the foregoing method, the management module 2 may set each virtual machine to have an equal maximum number of users and segmentation capacity; an exemplary manner in which the management module 2 configures the online user capacity of the plurality of virtual machines is: The maximum user value is set to m, and the value of the load sharing parameter P for initializing each virtual machine is n, where m and n are natural numbers, and the number of online users that can be accommodated in each of the segment capacities is m/n; An example of configuring a user to go online is to set the virtual machine whose current sharing parameter P is the largest as the outbound interface of the flow table. The user is online on the virtual machine until the number of online users of the virtual machine reaches (n-P+1). At the time of m/n, the value of the virtual machine sharing parameter P is decreased by one; the step is repeated until all the virtual machine sharing parameters P are 0 or all users are online; when multiple virtual machines are located When the shared parameter P is the maximum value, the virtual machine with the largest IP address is set as the outbound interface of the flow table, and the user is put on the virtual machine; when the shared parameter P of all the virtual machines is 0, the automatic virtual machine is automatically Virtual machine module 4 A new virtual machine is created in which the user is brought online in the new virtual machine.

The process of implementing multi-virtual machine load sharing by using the vBNG system device provided by the embodiment of the present disclosure is as follows: the user is deployed on the traditional router and the virtual device, and the parameters are set according to requirements, including the user capacity of each virtual machine (the most The number of large users, that is, the initial load sharing parameter P, after setting the parameters, starts the online user; the user sharing between the virtual machines is performed according to the steps shown in FIG. 3:

(1) The maximum number of users of each virtual machine in the management module is m;

(2) The management module initializes each virtual machine load sharing parameter P to n;

(3) The management module compares the load sharing parameter P, selects the virtual machine with the highest P value to write the flow table as the flow table outbound interface; if the P value is the same, compares the ip address, and selects the virtual machine with the ip address to write the flow table. As the flow table output interface;

(4) The management module records the number of online users of the current virtual machine, and compares the number of users with the standard value. The initial value of the standard value is m/n; if it is less than the standard value, it continues to go online on the virtual machine; Otherwise, the value of the load sharing parameter P of the virtual machine is decremented by 1, and the foregoing step 3 is repeated to select to switch to another virtual machine to continue the online user;

(5) When the number of online users of all virtual machines reaches m/n, the second round of user online is started from the third step, and the standard value in step 4 is increased by m/n, that is, the virtual machine is online. The number of users is compared with 2m/n; and so on, in each round, the user's online standard value is increased by m/n, and the load sharing parameter P value of each virtual machine is decremented by 1 until the load sharing parameter becomes 0;

(6) When the load sharing parameter P of all the virtual machines is 0, the virtual machine is automatically expanded to establish a new virtual machine.

The vBNG system device can adjust the virtual machine processed by the subsequent user in time according to the number of processed users on each virtual machine to realize rational utilization of resources; the routing device can usually use a traditional router, and the management module, the load sharing module, and the Virtual machine modules can be virtualized and work as virtual machines on a server that acts as a hardware carrier.

The embodiments of the above-mentioned vBNG system are schematic, and the common sense of the known structures and characteristics are not described in detail; the embodiments are described in a progressive manner, and the same similar parts between the embodiments can be referred to each other, and the embodiments are The technical features involved in the present invention may be combined with each other without conflicting with each other; wherein the units described as discrete components may or may not be physically separated, and the components presented as one unit may be physically The same place may be distributed in multiple networks; some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those skilled in the art can understand and implement the technical solutions of the present disclosure without any creative work.

In the description of the above-described embodiments of the present disclosure, the terms "first", "second", and the like are used for descriptive purposes only, and should not be construed as indicating or implying relative importance.

It should be noted that a number of improvements and modifications may be made by those skilled in the art without departing from the principles of the present disclosure, and such modifications and refinements are also considered to fall within the scope of the present disclosure.

Industrial applicability

The multi-virtual machine load sharing method of the vBNG system provided by the embodiment of the present disclosure can effectively adjust the virtual machine processed by the subsequent user according to the number of processed users on each virtual machine, which can effectively multi-multi-virtual machine vBNG system The number of users processed by the virtual machine is shared, and the virtual machines required for subsequent user packet processing are adjusted in time to realize rational utilization of resources, which can effectively reduce system instability caused by frequent switching of virtual machines.

Claims (15)

A vBNG system multi-virtual machine load sharing method is used for a vBNG network of a user online class, wherein the method includes: Receiving user protocol packets; Configure the online user capacity of each virtual machine, including setting the maximum number of users for each virtual machine, dividing the total user capacity of a single virtual machine into a number of segment sizes, and setting the online users that can be accommodated by each segment capacity of a single virtual machine. The sum of the numbers is equal to the maximum number of users of the virtual machine, and the number of segmented capacities of each virtual machine is set to be equal; Configuring each virtual machine to share the user's online connection, including sending the user protocol packet to the corresponding virtual machine for processing, so that the online user in the single virtual machine fills the segment capacity of the virtual machine one by one, and keeps the virtual machine in the virtual machine. The difference in the number of segmented capacities filled by the online user does not exceed one. The method of claim 1 wherein the method further comprises: When the number of online users in each virtual machine reaches the maximum number of users, a new virtual machine is automatically created. The method according to claim 1 or 2, wherein the maximum number of users of the virtual machines is set to be equal, and the segment sizes are equal in size. The method of claim 3, wherein the step of configuring an online user capacity of each virtual machine comprises: The maximum number of users of each virtual machine is set to m, and the value of the load sharing parameter P of each virtual machine is initialized, where m and n are natural numbers, and the number of online users that can be accommodated in each of the segmented capacities is m. /n. The method of claim 4, wherein the step of configuring each virtual machine to share the user's going online includes: Setting a virtual machine whose current load sharing parameter P is the largest in the flow table as an outgoing interface of the user protocol packet, and sending the user protocol packet to the virtual machine according to the flow table for processing, the virtual machine When the number of users reaches (n-P+1)·m/n, the user is stopped on the virtual machine, and the value of the virtual machine load sharing parameter P is decreased by one. This step is repeated until the load sharing parameter P of all virtual machines is repeated. 0 or all users are online. The method according to claim 5, wherein when the load sharing parameter P of the plurality of virtual machines is the maximum value, the user is brought online in a virtual machine in which the ip address is the largest. The method of claim 5, further comprising the steps of: When the load sharing parameter P of all virtual machines is 0, a new virtual machine is automatically created, and the user is brought online in the new virtual machine. A vBNG system device includes a virtual machine module, where the virtual machine module includes a virtual machine, wherein the device further includes a management module and a load sharing module, where: The management module is configured to generate a flow table required for the user to share the forwarding, and record the number of online users of each virtual machine in the virtual machine module and update the flow table; The management module is further configured to configure an online user capacity of each of the virtual machines, including setting a maximum number of users of each virtual machine, dividing a total user capacity of the single virtual machine into a plurality of segment capacities, and setting a single virtual Each of the machines The sum of the number of online users that can be accommodated by the segmentation capacity is equal to the maximum number of users of the virtual machine, and the number of segmentation capacities of each virtual machine is set to be equal; The management module is further configured to configure a user online policy, so that the online users in the single virtual machine fill the segment capacity of the virtual machine one by one, and keep the segments of each virtual machine filled by the online users. The difference in the number of capacities does not exceed 1; The load sharing module is configured to distribute user protocol packets to the virtual machine module according to the flow table and the user online policy. The device according to claim 8, wherein the management module is further configured to automatically establish a new virtual machine when the number of online users in each of the existing virtual machines reaches a maximum number of users. The device according to claim 8 or 9, wherein the maximum number of users of each of the virtual machines set by the management module is equal, and the segment sizes are equal in size. The device according to claim 10, wherein the configuring the online user capacity of each of the virtual machines of the management module comprises: setting a maximum user value of each of the virtual machines to m, and initializing each virtual machine The value of the load sharing parameter P is n, where m and n are both natural numbers, and the number of online users that can be accommodated in each of the segment sizes is m/n. The device of claim 11, wherein the step of configuring the user online policy of the management module comprises: setting a virtual machine with the current load sharing parameter P as the outbound interface of the user protocol packet in the flow table. According to the flow table, the user is online on the virtual machine until the number of online users of the virtual machine reaches (n-P+1)·m/n, and the value of the virtual machine load sharing parameter P is decreased by 1 at this time. Repeat the above steps until the load sharing parameter P of all virtual machines is 0 or all users are online. The device according to claim 12, wherein the configuration user online policy of the management module further comprises: when the load sharing parameter P of the plurality of virtual machines is the maximum value, the ip address is maximized A virtual machine is set as the flow table outbound interface, and the user is put on the virtual machine. The device according to claim 12, wherein the configuration user online policy of the management module further comprises: automatically creating a new one in the virtual machine module when the load sharing parameter P of all the virtual machines is 0 The virtual machine puts the user online in the new virtual machine. A storage medium, configured to store a program code, the program code for performing the vBNG system multi-virtual machine load sharing method according to any one of claims 1 to 7.

Images