KR20120063662A - Load balancer and method for managing load balancing using this - Google Patents

Abstract

PURPOSE: A load balancer and load distribution management method using the same are provided to offer the continuity of a service by automatically distributing communication service bottle neck sections. CONSTITUTION: A load balancer(10) constitutes a communication service cluster by allocating a VM(Virtual Machine) to physical nodes(50~53). The load balancer reconstitutes the communication service cluster by moving the VM according to states of the physical nodes corresponding to a communication service request received from clients. The load balancer determines the states of the physical nodes by monitoring the states in real time.

Description

Load balancer and method for managing load balancing using this}

The present invention relates to a load balancer and a load balancing management method using the same, and more particularly, to assign a virtual machine for each communication service to a physical node, tie the allocated virtual machines into a cluster for each communication service, Multiple virtual machines can be allocated redundantly to ensure service stability. Also, by dynamically allocating or migrating virtual machines according to the state of physical nodes, clusters can be reconfigured for communication service bottlenecks. The present invention relates to a load balancer capable of distributing and providing continuity of services and a load balancing management method using the same.

Existing load balancing technology uses a method of adding physical physical machines as backup equipment or configuring server clusters to secure service stability. When the service performance decreases due to a bottleneck, the conventional technology changes the internal structure so that the module can be parallelized by searching for a bottleneck section or finding a parallel processing section, or by adding a physical machine to increase the absolute amount of resources. I will try to solve by increasing. However, in such a case, a large cost and effort are required due to the change of software and the expansion of physical machines to change the internal structure of the service.

In addition, the existing load balancing technology uses a method of load balancing by introducing a load balancer (Load Balancer) to secure the scalability and stability of the server. The load balancer receives the packets sent to the website and sends them to each server based on one or more parameters. This load balancer selects the server to which the packet is sent in a way that distributes the packet load substantially evenly among the servers hosting the website. In addition, the load balancer load balances multiple instances in a cloud environment while providing STICKY SESSION management. Steam sessions can handle services while maintaining session information by forwarding all requests from one web server to the instance assigned to handle the first request when multiple requests arrive at the load balancer. Technology However, the session maintenance function applied to these load balancers can handle only HTTP and TCP protocols, and there is no consideration of other protocol processing.

In order to solve the above-mentioned problems, the present invention is to configure the virtual machines distributed in the physical node in the virtualization-based system in the virtualization-based system to secure the stability of the communication service and to eliminate the bottlenecks, the state of the physical node Accordingly, an object of the present invention is to provide a load balancer capable of dynamically allocating or migrating a virtual machine and a load balancing management method using the same.

Another object of the present invention is to provide a load balancer capable of performing a load balancing function while maintaining session information of various protocols (SIP, DIAMETER, MSRP, etc.) and a load balancing management method using the same.

An embodiment of the present invention for achieving the above object is to allocate a virtual machine to a plurality of physical nodes, bundle at least two or more of the allocated virtual machines for each communication service to form a communication service cluster, wired and wireless communication A load balancer that provides the requested communication service by reconfiguring a corresponding communication service cluster by allocating or migrating a new virtual machine according to the state of the plurality of physical nodes in response to a communication service request received from a client connected through the same. To provide.

On the other hand, another embodiment of the present invention for achieving the above object, the first to allocate a virtual machine for each communication service to a plurality of physical nodes, bundle the assigned virtual machine for each communication service to form a communication service cluster step; Reconfiguring a corresponding communication service cluster by creating or migrating a virtual machine according to a state of a physical node providing a requested communication service in response to a communication service request received from a client connected through wired or wireless communication; And a third step of providing a communication service requested by the client through a reconfigured communication service cluster.

According to the present invention, in a load balancer, a virtual machine is allocated to be distributed over a plurality of physical nodes, and a plurality of virtual machines are provided in a cluster by grouping them into communication services. As a result, it is possible to minimize the scope of service interrupted even in the event of a failure of a specific physical machine and service downtime, thereby ensuring the stability of the communication service.

In addition, according to the present invention, the load balancer can perform service load distribution of a virtual machine unit in a preconfigured virtual resource pool without changing the internal structure of the system. As a result, service load can be distributed without burdening the cost and effort, and bottlenecks can be eliminated.

In addition, according to the present invention, the load balancer may provide a load balancing function while maintaining session information of various protocols (SIP, DIAMETER, MSRP, etc.).

1 is a block diagram schematically illustrating a relationship between a plurality of clients and a web server that can be connected to each other around a load balancer according to an embodiment of the present invention.
2 is a block diagram showing the detailed configuration of a load balancer and a communication service cluster according to an embodiment of the present invention.
3 is a block diagram illustrating a load balancing concept for a bottleneck section of a load balancer according to an embodiment of the present invention.
4 is a block diagram illustrating a load balancing concept when a physical node of a load balancer fails in accordance with an embodiment of the present invention.
5 is an exemplary diagram illustrating a process of providing a communication service when a virtual machine failure of a specific physical node occurs in an IMS system using the load balancer illustrated in FIGS. 1 to 4.
6 is a flowchart illustrating a load balancing management method of a load balancer according to an embodiment of the present invention.
7 is a flowchart illustrating a load balancing management method for each protocol of a load balancer according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Further, in order to clearly illustrate the present invention in the drawings, portions not related to the description are omitted. In addition, the same code | symbol is attached | subjected about the similar part throughout the specification. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

The load balancer of the present invention and the load balancing management method using the same have a structure in which a resource is allocated to provide a communication service to an existing physical machine in a flexible structure based on a virtual machine.

That is, the load balancer of the present invention and the load balancing management method using the same, allocating a virtual machine for each communication service to a plurality of physical nodes, and comprises a cluster that provides a single communication service by tying a plurality of allocated virtual machines, By dynamically allocating or migrating a virtual machine according to the state of the physical node to reconfigure the cluster, it is possible to secure stability and continuity for communication services and to eliminate bottlenecks.

In addition, the load balancer of the present invention and a load balancing management method using the same may perform load balancing while maintaining sessions for various protocols such as HTTP, TCP, SIP, DIAMETER, MSRP, and the like.

For reference, in the specification of the present invention, the service and the communication service, the cluster and the communication service cluster are used in the same meaning, respectively, and may be used interchangeably.

1 is a block diagram schematically illustrating a relationship between a plurality of clients and an IMS-AS (Application Server) that can be connected to each other around a load balancer according to an embodiment of the present invention.

As shown in FIG. 1, the overall load balancing management system according to the present embodiment includes one load balancer 10, a plurality of clients 1-1, 30, and an IMS-AS 1-n 40. It includes.

Here, the load balancer 10 provides one access point for the client 30 and the IMS-AS 40 to provide a communication service according to the request of the client 30.

The load balancer 10 provides a cluster 20 for each communication service (1 to m). The communication service cluster 20 is a communication service providing unit configured to bundle virtual machines allocated to a plurality of physical nodes for each communication service.

In addition, the load balancer 10 manages cluster information configured for each communication service, and the cluster information includes virtual machine information and physical node information to which the virtual machine is allocated.

On the other hand, the load balancer 10 reconfigures the communication service cluster according to the state of the physical node. For example, if the state of the physical node is determined to be overloaded, create a new virtual machine on another physical node, or if the state of the physical node is determined to fail, migrate the virtual machine assigned to that physical node to another physical node. do.

To this end, the load balancer 10 manages the state of the physical node and the virtual machine assigned to the physical node, and determines the failure and overload of the communication service for the physical node. The load balancer 10 reconfigures the cluster according to the determination result of the state of the physical node, and changes the traffic distribution policy according to the reconfiguration of the cluster. At this time, the load balancer 10 maintains a session for each communication service for a communication service and distributes traffic to virtual machines allocated to physical nodes.

The load balancer 10 maintains session information (destination information) when the location of the virtual machine is changed by the migration of the virtual machine, and maintains sessions for each protocol such as HTTP, TCP, SIP, DIAMETER, and MSRP. Forward the message to the virtual machine.

At this time, the load balancer 10 manages the session information and the virtual machine information for processing the session, and requests the processing of the session to the virtual machine through the mapping of the session information and the virtual machine information.

As such, the load balancer according to the present invention can minimize the scope and downtime of the communication service which is stopped even when an overload or failure of a specific physical node occurs through a cluster structure of a virtual machine. As a result, the load balancer according to the present embodiment can secure stability and continuity for the communication service and eliminate bottlenecks.

2 is a block diagram showing the detailed configuration of a load balancer and a communication service cluster according to an embodiment of the present invention.

As shown in FIG. 2, the load balancer 10 according to the present embodiment includes a cluster manager 11, a traffic manager 12, a policy manager 13, and a session manager 14.

First, the cluster manager 11 allocates a virtual machine 60 to a plurality of physical nodes 50 to 53, and configures a communication service cluster 21 to 24 including a plurality of virtual machines for each communication service.

That is, the cluster manager 11 configures the virtual machines 60 distributed in the plurality of physical nodes 50 to 53 by one communication service cluster 21 to 24, respectively.

For example, the cluster manager 11 is a cluster 21 that bundles the virtual machines 60 distributed in the physical node 1 50 and the physical node 2 51 to provide communication service 1, and the physical node 2 ( 51), a cluster 22 which bundles the virtual machines 60 distributed in physical nodes 3 52 and 4 4 53 to provide communication service 2, wherein physical node 1 50 and physical node 3 ( 52 is a cluster 23 that bundles the virtual machines 60 distributed in a distributed manner to provide communication services 3, and includes physical node 1 (50), physical node 2 (51), physical node 3 (52), and physical node 4 ( 53, the virtual machines 60 distributed in a bundle are bundled together to form a cluster 24 that provides a communication service 4.

As such, the cluster manager 11 forms four communication service clusters on the plurality of physical nodes and registers and manages cluster information.

In addition, the cluster manager 11 changes the number of virtual machines included in the cluster according to the state of the physical node.

That is, when it is determined that a specific physical node is a service bottleneck, the cluster manager 11 distributes load by adding a virtual machine to the cluster and changes the cluster information.

In addition, the cluster manager 11 supports the movement between the physical nodes of the virtual machine according to the state of the physical node.

That is, if it is determined that the virtual machine of the communication service cluster is out of service due to a failure of a specific physical node, the cluster manager 11 migrates the virtual machine to another physical node, and migrates the physical node to which the virtual machine is migrated. Include in the communication service cluster and change cluster information.

In addition, the traffic management unit 12 manages the connection between the physical node and the virtual machine in order to perform traffic distribution according to the communication service, and determines whether the service is interrupted or overloaded.

In addition, the traffic management unit 12 distributes traffic to the corresponding virtual machine while maintaining the session and session information for each protocol, such as HTTP, TCP, SIP, DIAMETER, MSRP.

At this time, the traffic manager 12 performs traffic distribution according to the set traffic distribution policy for each protocol.

In addition, the traffic manager 12 notifies the cluster manager 11 when a specific physical node fails, and temporarily sets a traffic distribution policy by using a destination change function while the cluster manager 11 performs the migration of the virtual machine. After the change and migration of the virtual machine is completed, restore the traffic distribution policy before failure.

The policy manager 13 sets and manages a policy for distributing traffic to the virtual machine.

That is, the policy manager 13 sets the traffic distribution policy by reflecting the change of the location of the virtual machine according to the new creation and migration of the virtual machine in the event of service failure or overload.

In addition, the session manager 14 maintains session and session information and virtual machine information so that traffic distribution is seamless.

That is, the session manager 14 supports continuity of traffic distribution by maintaining session and session information for each protocol and then determining whether to maintain a session for a protocol input to the load balancer 10.

Next, the load balancing of the load balancer described above with reference to FIGS. 1 and 2 will be described in detail. 3 is a diagram illustrating load balancing for a bottleneck section of a load balancer, and FIG. 4 is a diagram illustrating load balancing when a physical node failure of the load balancer occurs.

3 is a block diagram illustrating a load balancing concept for a bottleneck section of a load balancer according to an embodiment of the present invention.

As shown in FIG. 3, the cluster manager 11 configures the virtual machine 61 of the physical node 1 50 as a cluster 21 for providing communication service 1.

The traffic manager 12 monitors the state of the physical node in real time, and notifies the cluster manager 11 when the performance decrease is confirmed as a service bottleneck for the physical node 1 50.

Then, the cluster manager 11 adds a new virtual machine 70 to the physical node 2 51, the physical node 3 52, and the physical node 4 53, respectively, and transfers the new virtual machines 70 to the communication service 1. Change the cluster information to be included in the cluster 21 to provide.

As a result, the cluster manager 11 distributes the load on the virtual machine 61 of the physical node 1 50 to the new virtual machine 70 to eliminate the bottleneck.

As such, the cluster management unit 11 of the load balancer may virtualize the cluster in order to solve this problem when the load on some areas of the processing logic of the communication service 1 degrades the performance of the entire system, thereby increasing the service processing time. By adding machines, the load can be distributed among bottlenecks, reducing overall latency.

At this time, the policy manager 13 of the load balancer changes the policy to distribute the load to the new virtual machine 70 as the virtual machine is added to the cluster 21 of the communication service 1.

Therefore, as described above, the load balancer according to the present embodiment can automatically distribute loads to service bottlenecks through new creation of the virtual machine.

4 is a block diagram illustrating a load balancing concept when a physical node of a load balancer fails in accordance with an embodiment of the present invention.

As shown in FIG. 4, the cluster manager 11 of the load balancer 10 includes at least two virtual machines processing four communication services on each of the four physical nodes 1 to 4 and 50 to 53. The clusters 21 to 24 are formed by overlapping the above steps.

Through this, the load balancer according to the present embodiment can secure the service stability by arranging virtual machines providing one communication service in a cluster.

If four communication services 1 to 4 are allocated and serviced to each of the four physical nodes 1 to 4, that is, communication service 1 to physical node 1 and communication service 2 to physical node 2, If communication service 3 is assigned to physical node 3 and communication service 4 is assigned to physical node 4, when the failure of physical node 2 occurs, communication service 2 assigned to it shows a 100% loss rate and cannot operate until the failure is recovered. If the communication service 1 to the communication service 4 operates in association with each other, it may cause a situation in which all communication services do not operate due to a failure of the physical node 2.

Thereafter, the traffic manager 12 monitors the states of the physical nodes 50 to 53 in real time, and when the failure of the physical node 2 51 is confirmed, the traffic manager 12 notifies the cluster manager 11.

Then, the cluster manager 11 migrates the virtual machine 62 of the failed physical node 2 51 to another physical node, physical node 1 50 and physical node 3 52, respectively.

That is, when a failure occurs in the physical node 2 51, the operation of the communication service 1 allocated to the physical node 2 51 and the virtual machine 62 of the communication service 2 is stopped, so that the communication service 1 is 50% and the communication service 2 is Although the loss rate is 33%, virtual machines distributed on different physical nodes continue to operate, so there is no interruption of the overall service.

Therefore, the cluster management unit 11 partially migrates the virtual machines assigned to the physical node 2 51 to the physical node 1 50 and the physical node 3 52, respectively, to cause failure of the physical node 2 51. Quickly resume service.

In this case, the load balancer 10 may continuously provide a service in a state before a failure occurs in the migrated virtual machine by using a session maintenance function and a protoco-specific processing function.

5 is an exemplary diagram illustrating a process of providing a communication service when a virtual machine failure of a specific physical node occurs in an IMS system using the load balancer illustrated in FIGS. 1 to 4.

As shown in Figure 5, the wireless communication terminal transmits the INVITE message to the load balancer (①).

Then, the load balancer selects the virtual machine of physical node 2 capable of processing the communication service (P-CSCF) requested by the IMS network, transmits an INVITE message to the virtual machine (②), and stores session information.

Then, the virtual machine of physical node 2 transmits the INVITE message to the PC according to the destination information (③).

At this time, when the failure of the physical node 2 occurs and the operation of the virtual machine assigned to the physical node 2 is stopped (④), the migration of the stopped virtual machine from the load balancer to the physical node 3 is determined, and physical / virtual resource management is performed accordingly. The virtual machine that is stopped in the system (not shown) is migrated to physical node 3 (⑤).

As a result, the virtual machine operates again on physical node 3, and the load balancer changes session information according to the migration of the virtual machine.

Thereafter, a 200 OK message is sent to the load balancer in response to the INVITE message from the PC (6).

Then, the load balancer compares the session information with the 200 OK message and migrates to physical node 3 to deliver the 200 OK message to the newly applied virtual machine (⑦).

Then, the virtual machine of physical node 3 transmits a 200 OK message to the load balancer, and the load balancer delivers a 200 OK message to the corresponding wireless communication terminal (⑧).

6 is a flowchart illustrating a load balancing method of a load balancer according to an embodiment of the present invention.

As shown in FIG. 6, the load balancer monitors the states of the plurality of physical nodes (S601).

As a result of the monitoring, when the state of the physical node is determined to be a performance degradation due to a bottleneck (S602, Y), it is checked whether the virtual machine can be allocated to another physical node (S603).

As a result of the check, if the virtual machine can be allocated (S63, Y), the new virtual machine is allocated to another physical node (S605).

Then, the load balancer changes the communication service cluster information (S605), and connects the new virtual machine and the load balancer (S606).

Then, the load balancer determines the traffic distribution position for each virtual machine (S607), and allocates a processing module for each protocol to the new virtual machine (S608).

Then, the load balancer distributes the traffic according to the communication service to the corresponding virtual machine (S609).

On the other hand, if it is determined that the allocation of the virtual machine is impossible (S603, N), the administrator is notified of the impossibility of changing the cluster configuration (S610).

On the other hand, if it is determined that the state of the physical node is a failure as a result of the monitoring (S611, Y), the virtual machine that has stopped operation due to the failure of the physical node is migrated to another physical node (S612).

At this time, the load balancer changes the traffic distribution location during the migration of the virtual machine (S613), and changes the communication service cluster information to another physical node migrated by the virtual machine (S614).

Then, the load balancer restores the traffic distribution policy before migration of the virtual machine (S615).

Then, the load balancer distributes the traffic according to the communication service to the corresponding virtual machine (S609).

7 is a flowchart illustrating a load balancing method for each protocol of a load balancer according to an embodiment of the present invention. In this embodiment, the case of SIP and MSRP is described as an example, but is not limited thereto.

As shown in FIG. 7, the load balancer analyzes a message transmitted for traffic distribution for each protocol (S701).

As a result of the analysis, if the message transmitted to the load balancer is a SIP message (S702, YES), the load balancer delivers the message to the SIP processing module (S703).

The load balancer analyzes the SIP message to determine whether to maintain session information (URI, TAG, CALL ID, CALL SEQUENCE information) (S704).

As a result of the check, if the session information is in the maintenance state (S704, YES), the load balancer maintains the session and selects the allocated virtual machine corresponding to the session information (S705).

Then, the load balancer temporarily stores the transmitted SIP message and rearranges the order of the SIP message according to the CALL SEQUENCE (S706).

The load balancer transfers the SIP message to the selected virtual machine (S707).

On the other hand, if the session information is not in the maintenance state as a result of the check (S704, NO), the communication service cluster information is inquired, the virtual machine capable of providing the corresponding communication service is allocated, and the load balancer and the distributed virtual machine are connected (S708). .

Then, the load balancer matches and stores the session information of the transmitted SIP message and the allocated virtual machine information (S709).

The load balancer transfers the SIP message to the distributed virtual machine (S707).

On the other hand, if the analysis result, the message sent to the load balancer MSRP message (S710, YES), the load balancer delivers the message to the MSRP processing module (S711).

In addition, the load balancer analyzes the MSRP message and checks whether matching information about the calling terminal address and session information (SESSION ID) is registered (S712).

As a result of the check, if the SESSION ID for the calling terminal address is registered (S712, YES), the virtual machine assigned to the SESSION ID is selected (S713), and the MSRP message transmitted to the selected virtual machine is transmitted (S707).

On the other hand, if the SESSION ID for the calling terminal address is not registered (S712, NO), the load balancer allocates a new SESSION ID (S714).

Then, the load balancer allocates a virtual machine to process a service corresponding to the allocated new SESSION ID, and stores allocation information of the virtual machine for the SESSION ID (S715).

The load balancer transfers the MSRP message to the allocated virtual machine (S707).

So far I looked at the center of the preferred embodiment for the present invention. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments are to be regarded as illustrative rather than restrictive, and all differences within the equivalent scope should be construed as being included in the present invention.

Further, the system and method according to the present invention can be embodied as computer readable codes on a computer readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the recording medium include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like, and a carrier wave (for example, transmission via the Internet). The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

10. Load Balancer 11. Cluster Management
12. Traffic Management Unit 13. Policy Management Unit
14. Session Manager 20 ~ 24. Communications Service Cluster

Claims (19)

Assign a virtual machine to a plurality of physical nodes, bundle at least two or more allocated virtual machines by communication service to form a communication service cluster,
In response to a communication service request received from a client connected through wired / wireless communication, the requested communication service is provided by reconfiguring a corresponding communication service cluster by allocating or migrating a new virtual machine according to the state of the plurality of physical nodes. Load balancer. The method of claim 1,
The load balancer determines the state of the plurality of physical nodes through real-time monitoring, and if the physical node that provides the requested communication service is a bottleneck, the bottleneck is allocated to another physical node by assigning a new virtual machine to the bottleneck. The load balancer, characterized in that for changing the included communication service cluster information. The method of claim 2,
The load balancer connects the allocated new virtual machine and the load balancer, determines a traffic distribution location for the virtual machine of the changed communication service cluster, and allocates a processing module for each protocol to provide the requested communication service. Load balancer. The method of claim 1,
The load balancer determines the state of the plurality of physical nodes through real-time monitoring, and if it is determined that the physical node providing the requested communication service is a failure section, the load balancer may change the virtual machine of the physical node that is the failure section. And migrating to a physical node and changing communication service cluster information including the failure section. The method of claim 4, wherein
The load balancer changes the traffic distribution location during the migration of the virtual machine, and restores the traffic distribution policy before migration of the virtual machine to provide the requested communication service when the migration of the virtual machine is completed. Featuring a load balancer. The method of claim 1,
The load balancer includes a protocol-specific processing module including HTTP, TCP, SIP, DIAMETER, and MSRP, and load balancer for performing traffic distribution by maintaining session and session information for a corresponding protocol. A first step of allocating a virtual machine for each communication service to a plurality of physical nodes and bundling the allocated virtual machines for each communication service to form a communication service cluster;
Reconfiguring a corresponding communication service cluster by creating or migrating a virtual machine according to a state of a physical node providing a requested communication service in response to a communication service request received from a client connected through wired or wireless communication; And
And a third step of providing a communication service requested by the client through a reconfigured communication service cluster. The method of claim 7, wherein
The second step may include: real time monitoring a state of the plurality of physical nodes;
Determining whether the state of the physical node providing the requested communication service is a performance degradation or failure due to an overload; And
And reconfiguring the corresponding communication service cluster according to the determination result. The method of claim 8,
The reconfiguring of the corresponding communication service cluster may include: if the state of the physical node providing the requested communication service is degraded due to an overload, as a result of the determination, the virtual allocated to the physical node providing the requested communication service Determining a machine as a bottleneck, and allocating a new virtual machine to another physical node; and load balancing management method of a load balancer. 10. The method of claim 9,
Reconfiguring the corresponding communication service cluster,
Acquiring a distribution policy of a virtual machine if the state of the physical node providing the requested communication service is degraded due to the result of the determination;
Assigning a new virtual machine to at least one other physical node;
Changing the corresponding communication service cluster information;
Connecting the new virtual machine and a load balancer;
Determining a traffic distribution location for each new virtual machine; And
Allocating a processing module for each protocol to the new virtual machine; Load balancing management method of the load balancer comprising a. According to claim 8,
Reconfiguring the corresponding communication service cluster may include migrating a virtual machine of the failed physical node to another physical node if the state of the physical node providing the requested communication service is a failure as a result of the determination; Load balance management method of a load balancer, characterized in that. The method of claim 11,
Reconfiguring the corresponding communication service cluster,
Changing the traffic distribution location of the corresponding communication service cluster while migrating the virtual machine of the failed physical node to the other physical node;
Changing the corresponding communication service cluster information to the other physical node; And
Restoring the changed traffic distribution location before failure occurs; Load balancing management method of a load balancer, characterized in that it further comprises. The method of claim 7, wherein
The third step may further include performing traffic distribution for each communication service requested from a client connected through the wired / wireless communication.
The traffic distribution for each requested communication service is performed by maintaining session and session information for each protocol. The method of claim 13,
The third step, the load balancing management method of the load balancer, characterized in that the communication service is processed according to the matching information of the session information for each protocol and the virtual machine information processing the session. The method of claim 13,
Performing the traffic distribution for each of the requested communication services,
Analyzing the received message to determine a protocol type;
If the message is a SIP message, transmitting the SIP message to a SIP processing module;
Analyzing the transmitted SIP message to determine whether there is corresponding SIP session information;
Confirming that the SIP session information exists, selecting a virtual machine allocated to perform the SIP session;
Rearranging the message order according to call sequence information of the analyzed SIP message; And
And delivering the SIP message to the selected virtual machine. 16. The method of claim 15,
Allocating a virtual machine for providing a corresponding communication service if there is no SIP session information as a result of the checking;
Storing analysis information of the SIP message and allocation information of a virtual machine; And
And transmitting the received SIP message to the allocated virtual machine. 16. The method of claim 15,
If the message is an MSRP message, transmitting the MSRP message to an MSRP processing module;
Analyzing the transmitted MSRP message and confirming matching information between the calling terminal address and a session ID;
Selecting the virtual machine assigned to the session ID if the matching information is found as a result of the checking; And
Delivering the received MSRP message to the selected virtual machine; Load balancing management method of the load balancer further comprising. The method of claim 17,
If the matching information is not found, allocating a new session ID and allocating a virtual machine for providing a corresponding communication service;
Storing allocation information of the new session ID and the virtual machine; And
Delivering the received MSRP message to the assigned virtual machine; Load balancing management method of a load balancer further comprising. A computer-readable recording medium having recorded thereon a program for executing the method according to any one of claims 1 to 12.

Images