KR101678612B1 - Method and System for providing Dynamic Policy Application, Load-Balancing, and failover by dynamically controlling Proxy-Selector and Proxy configuration in each Managed End-Device - Google Patents

Abstract

The present invention relates a service system for providing functions of traffic load distribution, server duplication, and application of a network policy and service in a current network, and a method for controlling the service system. More specifically, the present invention relates to the service system in which proxy selectors of respective terminals and setting of proxies are controlled by a central controller and a specific traffic of a corresponding terminal is controlled to sequentially pass through a specific service server group or a plurality of service server groups, and thus a specific policy or service is enabled to be applied to the specific traffic, generated by the terminal, at any desirable time and specific network servers are enabled to provide services for load distribution and duplication at a desirable time in response to a service request. The present invention also relates to the method for controlling the service system.

Description

TECHNICAL FIELD [0001] The present invention relates to a service system providing load-balancing and failure recovery, and a control method thereof, -Selector and Proxy configuration in each Managed End-Device}

The present invention relates to a service system and a control method thereof for providing traffic load balancing, server redundancy, network policy, and service application functions in a current network, and more particularly, to a proxy system Proxy settings are controlled by a central controller so that the specific traffic of the terminal is sequentially passed through a specific service server group or a plurality of service server groups to allow application of a specific policy or service for specific traffic generated from the terminal at any time And to provide a load distribution and a duplication service to specific network servers at any time depending on whether or not the service application is made, and a control method thereof.

In a conventional load balancing and failure recovery technique, a service server group or a network function server group uses a load balancing function to allow a plurality of servers providing the same service to request a capacity of more than a single server can handle alone So that even when a failure occurs in a specific server of the server group, requests that have been processed by the failed server are distributed to the normal servers so that the failure can be quickly recovered. These load balancing and failover techniques are usually provided by installing hardware devices in front of the server group. Therefore, the servers that are provided with the load balancing and failback functions should be physically located close to each other, Hardware devices must be installed.

In addition to load balancing and failback through physical hardware devices, DNS (Domain Name Server) distributes the IP address of the server when requesting a server IP (Internet Protocol) address from the client based on the server load or failure of each server. And may provide a load balancing service. However, DNS using this method can provide only failover and load balancing to the final destination. Therefore, for a service server group providing differentiated services before a packet transmitted on the network reaches a final destination, And load balancing service is not possible, and in case of failure, it is necessary to go through the process of requesting and receiving the DNS again in order to recover the failure.

Proxy technology is a technique that allows a packet to be delivered to its final destination through a specific proxy server in transmitting the packet. This technique can potentially be used for a variety of purposes, but it is commonly used to lower the utilization of the external network by using the cache of the proxy server, to track the usage rate of the network, And to prevent viruses or malicious codes from propagating through packets. That is, there are advantages that the benefits of the clients using the proxy vary according to the services provided by the proxy server.

Meanwhile, conventional failover and load balancing require the server administrator to purchase, install, and maintain expensive load balancing hardware equipment. This problem can be solved to some extent by reducing the purchase demand of physical load balancing hardware equipment through load balancing and failback technology through DNS. However, load balancing and failback through DNS can only be performed by servers located in the final stage In the following, the load balancing and failure recovery are not supported for a service server group that directly intervenes in a communication line other than a separate root such as a router, a firewall, or an intrusion detection / There is a problem in that it takes time to request a new IP address to the DNS and to allocate it.

In addition, current industry trends that provide the necessary resources or services such as Platform as a Service (PaaS), Computer as a Service (CaaS), and Service as a Service (SaaS) can be used for router, firewall or intrusion detection / Service servers are encouraged to be provided as a service rather than a hardware device. In such environments, service servers providing differentiated processing services for the above packets are not the final destinations of the packets but are destinations that go through. Although not the final destination of such a packet, service server families that provide special handling for particular packets or packets transmitted from a particular user will inevitably need its load balancing and failover capabilities, It is required for technology that enables a software approach that can be applied flexibly according to demand, not equipment purchase.

Japanese Patent Application Laid-Open No. 2009-245309 : Access request transmission system, access request transmission method, and access request transmission program Korean Patent No. 10-1480128 : NDLP Appliance System Supports Mirroring and Inline Load Balancing and Redundancy, and Load Balancing and Redundancy Method Using It

SUMMARY OF THE INVENTION It is an object of the present invention to provide a service system and a control method thereof for providing real-time application of network policy, load balancing, and failure recovery.

In addition, the present invention provides a service server group that is differentiated according to each packet in an environment having a service server group that provides a differentiated network function for each packet flow at a certain position on the network by utilizing a proxy technology, The present invention provides a service system and a control method thereof that provide real-time application, load balancing, and failover of a network policy that can provide load balancing and failure recovery for each server of a group.

However, the object of the present invention is not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

A centralized policy server, called a central control device, controls a daemon for controlling proxy server settings of a proxy client installed in a terminal, so that a specific packet transmitted from the terminal can be passed through a specific service server. Policy is differentiated for each user and service, and the central control unit, which is a central policy server, checks the status of each service server in real time, so that the IP address of the proxy server set in each terminal proxy client may be different, So that it can support the load balancing and the failure recovery function for the service server.

A service system providing policy application, load balancing, and failure recovery, including a central control device, one or more terminals, and a plurality of internal server servers, according to an exemplary embodiment of the present invention, A differentiated policy can be applied to each packet by using a different proxy server for each packet generated by the terminal according to the rule and the proxy server IP of the terminal control rule that the central control device gives to the terminal is adjusted, The terminal can receive the list of the next proxy servers through the terminal control rules, so that even when a failure occurs in the currently connected proxy server, the terminal can immediately return to the next proxy server Providing fast failover capabilities to the proxy server by performing the connection .

Preferably, the central control unit calculates a terminal control rule based on a load degree and a failure degree of each internal server of the service server group, a preset terminal-specific policy, and a service server group policy, And distributes the load between the servers in the service server group and the failure recovery function of the internal server by providing different services to the packet generated by the terminal and transmitting the packet to the terminal.

Preferably, the at least one terminal establishes a terminal proxy selector, a proxy client, or a next proxy server list in accordance with the terminal control rules received from the central control device, so that packets generated by the terminal, as controlled by the central control device, The other service server group is passed through as a proxy server so that the differentiated service is applied to each packet, and the central control unit controls the proxy server set for each terminal to provide load balancing, The proxy server can be restored to its original state when a failure occurs in the proxy server.

More preferably, the proxy selector identifies a packet based on a part of identifiers set in the packet identification information that can be used for packet identification such as L2 to L7 packet information and terminal internal application identifier, and then transmits the packet to the set proxy To the client.

More preferably, the proxy client is connected to a proxy server supported by a server of the service server group among available proxy client types, and uses a proxy client type having a function necessary for allowing a packet to be received via the proxy server .

More preferably, the next proxy server list is a list for storing information of backup proxy servers to be newly used when a failure occurs in a proxy server set for each proxy client. The list is stored for each proxy client and is stored in the proxy server of the specific proxy client When a failure occurs, the terminal finds the backup proxy information immediately without consulting the central control device, and replaces the proxy server information with the backup proxy server to help complete the failure recovery. Also, the next proxy server list is a list, in which a plurality of backup proxy servers for the corresponding proxy client are provided as a list, so that the terminal can attempt connection to the backup server in the order of the list so that one or two backup servers So that the backup server can be found and the failure recovery can be completed. In addition, the servers included in the next proxy server list are already providing services to packets of other terminals, and when the central control device releases the terminal control rules, It is possible to use servers already in use without a standby server for restoration while providing a load distribution function for a backup server by providing the list of the next proxy servers having the servers in different orders.

Preferably, the service server group is a set of servers that provide services that are differentially provided for each terminal and each packet, and each service server group provides specific services specific to each policy, So that a set differential service can be provided.

A method of controlling a service system providing load balancing and failure recovery according to another embodiment of the present invention includes: detecting a load and a load of each internal server of a registered service server group periodically; A packet generating and transmitting step of processing a packet by itself in accordance with a set terminal control rule; Failure in case of failure of internal server of service server group Detection of failure and notification of failure notification that all the terminals that were connected to the server are notified of the failure or not, ; And a failure recovery step of completing a failure recovery by replacing the proxy server setting in which the terminal itself fails with a backup proxy server setting; And a control unit.

The control method includes registering a policy for receiving a service for each terminal or each packet, registering a service server group receiving a load balancing and failback service and registering internal servers of the server group .

Preferably, the failure and load detection step is a step in which the central control device collects load information and failure information from the service server group periodically applying the load balancing and / or failure recovery service through a predetermined protocol or program from the service server group .

In the control method, the central control device calculates a terminal control rule for each terminal, transmits the calculated terminal control rule to each terminal, processes each packet based on the terminal control rule set transmitted from each terminal, And a terminal control rule setting step of applying a policy differentiated according to a terminal and a packet and providing a load balancing and a failure recovery for a service server group.

The packet generation and delivery step is performed such that each terminal receives packets from the central control apparatus and processes packets generated by the terminal internal application according to a predetermined terminal control rule.

In the failure detection and failure notification step, the central control unit detects the failure and notifies the terminals connected to the failed server of the failure, or each terminal itself detects the failure, And the like.

The failure recovery step may be performed immediately after the terminal recognizes the failure by quickly replacing the failure proxy server setting with the proxy server for backup in the next proxy server list preset by receiving the terminal control rule .

A control method of a service system that provides load balancing and failure recovery, including a central control device, one or more terminals, and a plurality of internal server servers, according to another embodiment of the present invention, A policy applying step of allowing a policy differentiated according to a packet to be applied through different proxy servers for each packet generated by the terminal according to the policy; A load balancing step of controlling a proxy server IP of a terminal control rule provided to the terminal by the central control device so that the central control device is connected to a different proxy server for each terminal to perform a load balancing function; And a failure recovery step of receiving a list of the next proxy servers through the terminal control rules and performing reconnection to the next proxy server even when a failure occurs in the currently connected proxy server to provide a failure recovery function to the proxy server .

For example, the central control unit may calculate a terminal control rule based on a degree of load and a failure of each internal server of the service server group, a preset terminal policy and a service server group policy, To provide differentiated application of services to the packets generated by the terminal by differentiating the proxy server to be connected to each terminal by providing load balancing between the internal servers of the service server group and failure recovery function of the internal server .

The at least one terminal may set a terminal proxy selector, a proxy client, or a next proxy server list according to a terminal control rule received from the central control device, and the packets generated by the terminal are transmitted to different proxy servers The centralized control device can control the proxy server set for each terminal so as to provide differentiated services for each packet and to provide the load balancing. In addition, when a failure occurs in the proxy server through the list of the next proxy server preset in the terminal, So that recovery can be performed.

More preferably, the proxy selector identifies a packet on the basis of a part of identifiers set in the packet identification information that can be used for packet identification such as L2 to L7 packet information and terminal internal application identifier, To the proxy client set for the packet.

More preferably, the proxy client is connected to a proxy server supported by a server of the service server group among available proxy client types, and uses a type of proxy client having a function necessary for allowing a packet to be received via the proxy server The packet is set and transmitted so that the packet transmitted to the proxy client can pass through the proxy server connected to the proxy client.

For example, the next proxy server list is a list for storing information of a backup proxy server to be newly used when a failure occurs in the proxy server set for each proxy client. The list is stored for each proxy client so that a failure occurs in the proxy server of the specific proxy client The proxy server information of the corresponding proxy client is replaced with the backup proxy server immediately without consulting the central control device so that the failure recovery is completed and the server of the next proxy server list descending for each terminal is changed to perform the load balancing simultaneously with the load balancing And a service providing server can be set as a backup proxy server, thereby providing failover without a backup standby server.

For example, the service server group is a set of servers that provide services that are differentially provided for each terminal and each packet. Each service server group provides specific services specific to each policy, Server group so that different services can be applied on a packet-by-packet basis.

A computer-readable recording medium according to another embodiment of the present invention records a program for executing a control method of a service system that provides policy application, load balancing, and failure recovery.

As described above, using the service system that provides policy application, load balancing, and failure recovery of the present invention, the central control apparatus 100 exists on the network, and the terminal control module 201 installs It is possible to provide load balancing and fail-over functions for each server of all the service server groups 300 on the network without installing or removing new hardware, and it is cost effective and efficient.

Since the present invention can be applied regardless of the number of service servers, it has an advantage that a capacity expansion can be facilitated by adding a server.

The detection of the failure and the recovery of the failure can be performed quickly by setting the bypass path to the next proxy server for the failure recovery after the failure is detected actively in the terminal so that the load distribution between the servers can be performed while the failure recovery is in progress .

In addition, according to the present invention, a differentiated policy can be provided for each terminal in order to apply a terminal control rule to each terminal, and different policies can be applied for each traffic or terminal application through the proxy selector 212 in the terminal It is possible to provide a detailed service in a different manner.

Figure 1 illustrates a service system providing policy enforcement, load balancing, and failover in accordance with the present invention.
2 shows a central control device configuration according to the present invention.
FIG. 3 shows a configuration inside the terminal according to the present invention.
FIG. 4 is a flowchart illustrating a process of policy application and load balancing according to the present invention.
FIG. 5 illustrates a process of performing a failure recovery through a terminal control according to the present invention.
FIG. 6 illustrates a processing procedure of a control method of a service system that provides policy application, load balancing, and failure recovery according to another embodiment of the present invention.
FIG. 7 illustrates a processing procedure of a service system control method for providing policy application, load balancing, and failure recovery according to another embodiment of the present invention.
8 is a configuration example of a service system according to the present invention. In an environment where an MPTCP gateway capable of providing MPTCP performance to a final server 400 that can not use the MPTCP is used, Distributed, and disaster recovery.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention is capable of various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the term "comprising" or "comprising" or the like is intended to specify the presence of stated features, integers, But do not preclude the presence or addition of features, numbers, steps, operations, components, parts, or combinations thereof.

Figure 1 shows a service system providing policy enforcement, load balancing, and failover in accordance with the present invention. The present invention relates to a service system comprising a central control device (100), one or more terminals (200), and a service server group (300) composed of a plurality of internal servers and providing policy application, load balancing, to be. The service system creates and sets a terminal control rule for each terminal based on pre-registered policy and service server group information. In accordance with the terminal control rule, the terminal generates a control rule for each packet generated by the terminal through different proxy servers The central control apparatus 100 controls the proxy server IP of the terminal control rule for each terminal to be connected to a different proxy server for each terminal, The terminal receiving the next proxy server list through the terminal control rule can quickly reconnect to the next proxy server even when a failure occurs in the currently connected proxy server to quickly provide the failure recovery function to the proxy server . That is, the central control unit 100 calculates a terminal control rule based on the degree of load and failure of each internal server of the service server group 300, the preset terminal-specific policy and the service server group policy, The service server group 300 may be configured to provide a packet service application for a packet generated by the terminal by transferring the terminal control rule to the terminal, and to differentiate the proxy server to be connected to each terminal in the terminal control rule, Load balancing among internal servers, and failure recovery of internal servers.

The at least one terminal 200 may set a terminal internal proxy selector, a proxy client, or a next proxy server list according to the terminal control rules received from the central control apparatus 100, The generated packets are transmitted to different proxy servers so that differentiated services are applied to the packets and the central control device 100 can indirectly control the proxy server set for each terminal through the terminal control rules to provide load balancing And allows the proxy server to immediately recover from a failure in the event of a failure through the next proxy server list preset in the terminal. The proxy selector identifies a packet on the basis of a part of identifiers set in packet identification information that can be used for packet identification such as L2 to L7 packet information and terminal internal application identifier, It can be transmitted to the proxy client that has been set up with the proxy server according to the terminal control rule so as to receive the service to be provided to the packet. The proxy client is connected to a proxy server supported by the server of the service server group 300 among the available proxy clients and transmits the service provided by the central control apparatus 100 through the service server group 300 to the packet The proxy client can be used to set up and forward the packet so that the packet forwarded to the proxy client passes through the proxy server connected to the proxy client. Here, the optimal proxy client can be predetermined or selected by the administrator based on whether the available proxy client can satisfy the transmission quality of the service and / or the transmission quality such as transmission delay, packet error rate, and the like.

The next proxy server list is a list for storing information of a backup proxy server to be newly used when a failure occurs in a proxy server set for each proxy client. The list is stored for each proxy client, and when a failure occurs in a proxy server of a specific proxy client, The backup proxy information is immediately found without consulting the central control device, and the proxy server information is replaced with the backup proxy server to complete the failure recovery. Also, the next proxy server list is a list, in which a plurality of backup proxy servers for the corresponding proxy client are provided as a list, so that the terminal can attempt connection to the backup server in the order of the list so that one or two backup servers It is possible to find a backup server that operates even if a failure occurs and to complete the failure recovery. In addition, the servers included in the next proxy server list are already providing services to packets of other terminals, and when the central control device releases the terminal control rules, It provides the load distribution function for the backup server by giving the server to the next proxy server list in a different order so that the servers that are already in service can be used without a standby server (standby).

The service server group 300 is a set of servers that provide services that are differentially provided for each terminal and each packet. Each service server group 300 provides specific services specific to each policy, And / or receive differentiated services on a packet - by - packet basis.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The service system providing policy application, load balancing, and failure recovery according to the present invention includes a central control device 100, a terminal 200, and a service server group 300, as shown in FIG. The operation of this service system will be described with reference to Fig.

First, when the terminal 200 turns on to receive the service of the service system and starts connecting to the network, the terminal (e.g., terminal terminal setup agent) informs the central control apparatus 100 of the start of the terminal 200 And receives a terminal control rule transmitted from the central control apparatus 100. [

Next, the terminal 200 sets a proxy selector and a proxy client according to the terminal control rule, and stores the next proxy server list in the memory (not shown) of the terminal 200.

Next, when a packet is generated in the terminal 200, the proxy selector delivers a packet corresponding to the proxy client set for each packet, and each proxy client sets the received packet to pass through the set proxy server, To the network through which the service system of the present invention performs policy application and load balancing without purchasing a separate hardware device. The proxy selector identifies the packet based on the information that can be used for packet identification including the packet information of L2 to L7 and the internal application identifier of the terminal 200, and transfers the packet to the different proxy clients. At this time, the packet information of L2 to L7 and the internal application identifier of the terminal 200 are used as an example, and the present invention can use all identifiers which are not limited to the example identifiers and are capable of packet identification. Each proxy client plays a role of handling different service servers via a packet for each proxy client so that differentiated services can be applied to different packets in cooperation with different proxy servers.

When a failure occurs, the terminal 200 sets a next proxy server in the next proxy server list set for each proxy client after detecting the failure, to the proxy client so as to replace the proxy server information in which the problem occurs, thereby performing failover. The next proxy server list is one for each proxy client, and when a failure occurs, it stores the list to be connected to the next proxy server sequentially in memory.

The service server group 300 periodically transmits the severity and severity of the server to the central control unit 100 by using any of commonly known or specially manufactured protocols and programs. Different service server groups exist for each packet, and differentiated services can be provided. In the service server group, a plurality of servers exist so that each server can serve as a backup server of each other.

2 shows a central control device configuration according to the present invention. 2, the central control apparatus 100 includes a policy processor 101 and a terminal controller 104. [ The policy processor 101 may be implemented by including a policy searcher 102, a policy DB 103, and a policy generator 105. The operation of the central control unit 100 will be described below.

First, the central control device 100 periodically collects the degree of load and the failure of the server. The degree of load and fault information can be collected by using a conventional communication protocol (TCP / IP, UDP / IP, or the like) and programs, or by using a special communication protocol and program separately. The central control unit can generate a terminal control rule for each terminal based on the policy, the failure of each server, and the load of each server. The policy generator 105 generates a policy according to a user's input through a UI (User Interface). The policy DB 103 stores the generated policy, the failure status and load of each server, and the current terminal control rule set in each terminal 200. The policy searcher 102 calculates a terminal control rule for the corresponding terminal 200 based on the information of the policy DB 103. The terminal control rules are calculated and transmitted on demand basis (On-Demand). The calculation algorithm can be any known or known to the ordinary artisan. The terminal controller 104 transfers the terminal control rule in cooperation with the terminal setting agent 211 of the terminal 200 and updates the policy DB 103 to update the terminal control rule currently being used by the terminal 200 .

Next, the terminal control rule for each terminal is calculated when the terminal 200 is started, and is immediately transferred to the terminal to set the terminal.

In addition, the terminal 200 may transmit its terminal control rule information to the central control apparatus 100 at a set interval so that the central control apparatus 100 updates the set terminal control rule information stored for each terminal 200 .

The terminal control rule transmitted in this way controls the information of the proxy selector, the proxy client, and the next proxy server list of the terminal 200 to indirectly control the service provision for the packet generated by the terminal .

FIG. 3 shows the configuration of each terminal according to the present invention. As shown in FIG. 3, the terminal 200 includes a terminal control module 201, which includes a terminal configuration agent 211, a proxy selector 212, one or more proxy clients 213, And a next proxy server list 214. The operation of the terminal will be described below.

First, the terminal control agent 211 of the terminal control module 201 sets the proxy selector, the proxy client, and the next proxy server list using the information of the received terminal control rule. That is, the terminal control agent 211 of the terminal control module 201 notifies the central control device 100 of the start of the terminal when the terminal is started, receives the terminal control rule 202, The client 213, and the next proxy server list 214, respectively.

When the setting of the terminal control module 201 is completed, the packet generated by each application of the terminal 200 is received by the proxy selector 212 of the proxy client 213, As shown in FIG.

The proxy selector 212 identifies a packet or a packet source through L2 to L7 packet information and an application identifier. Set an action for the identified packet so that the packet is processed by that action. Typically, an action means to perform packet forwarding to a specific proxy client.

The proxy server setting is performed through the terminal control agent 211. The proxy client 213, in cooperation with the proxy server, sets the received packet via the corresponding proxy server, and then releases the packet to the network. That is, the proxy client 213 allows the received packet to go through the set proxy server, and then the packet is released to the network.

In the case of the proxy client 213, all the proxy clients that the terminal 200 and the service server can support are applicable, and the manner in which the packet can pass through the proxy server may be different for each proxy client used. The present invention does not specify which proxy client is used, and indicates it as "..." in FIG. The SOCKS proxy and HTTP proxy described in FIG. 3 are examples of the types of proxy clients that can be used.

The next proxy server list is a list that the terminal 200 receives from the central control device 100 for fast failure recovery. If a failure is detected in the proxy server set in each proxy client, the terminal itself can notify the next proxy server So that a quick failover can be performed.

There is one proxy server list for each proxy client, so that the proxy client performs fast failback without inquiring to the central control device separately. Proxy server for backup is kept in list form, so that even if several proxy servers for backup occur at the same time, it detects the proxy server in operation and performs failback. Since the server and server order in the next proxy server list that goes down through the UE control rules are different for each UE, the failure recovery is performed simultaneously with load balancing. Since the server registered in the next proxy server list can set a normal server that is currently in service, not a standby server for backup, which is not inactive (not activated), it is possible to perform a failure recovery without preparing a backup server.

Since one proxy client is linked with one service server group 300, the next proxy server list must also be received and stored for each proxy client.

The central control apparatus 100 of the present invention provides the IP address of the server to be set in the proxy client based on the load information of each server of the service server group 300 when releasing the terminal control rule. That is, the server having the lowest degree of load is included in the terminal control rule as the proxy client configuration information, or the selected server is down as the proxy server through the other load balancing algorithm. In this way, different servers are used as proxy servers for each terminal, thereby providing an equal load balancing function.

In this case, the load information may include a CPU usage amount, a memory usage amount, and a number of connected sessions, and the CPU usage amount, the memory usage amount, the number of connected sessions, and the like are the load information used in the present invention. The criteria of the load information to be used can be changed according to the application of the system of the present invention, and the central control apparatus 100 of the present invention requires the load information required for processing and the weight (weight, importance) As shown in FIG.

FIG. 4 illustrates an example of a process of performing load balancing through packet processing according to a UE control rule according to the present invention. As shown in FIG. 4, (1) the service server receiving the load balancing and failback service is registered by the manager 500. (2) When the client Client_1 is started, (3) Client_1 passes its start to the central control device 100, and then receives the terminal control rule including the proxy selector setting, the proxy client setting, and the list of the next proxy server. At this time, the central control device 100 recognizes that the servers ServerA and ServerB, which are servers in the service server group 300, are not loaded, transfers ServerA to use as a proxy server, and transmits ServerB to the next proxy server It is included in the proxy server list and transmitted. (4) The packet sent from Client_1 to Server0, which is the final server 400, is transmitted to the proxy client applied to the packet by the intra-terminal proxy selector, and is set to pass through the server A, which is the proxy server, and is released to the network. (5) Server A operates as a proxy server and delivers a packet to Server 0, which is the final server 400, after performing a service on a packet. (6) When the client Client_2 is started, (7) Client_2 transfers its start to the central control device 100 and receives the terminal control rule including the proxy selector setting, the proxy client setting, and the list of the next proxy server. At this time, since the central control device 100 recognizes that there is more load on the server A, the central control device 100 transmits the server B to use as a proxy server, and transmits the server A to the next proxy server list by including it in the next proxy server list. (8) The packet transmitted from Client_2 to Server0, which is the final server 400, is transmitted to the proxy client applied to the packet by the intra-terminal proxy selector, and is set to pass the ServerB, which is the proxy server, (9) Server B operates as a proxy server and delivers a packet to the final server 400 after performing a service for the packet.

FIG. 5 illustrates a process of performing a failure recovery through packet processing according to a UE control rule according to the present invention. As shown in FIG. 5, (1) when a failure occurs in the server A among the servers of the registered service server group 300, (2) the central control apparatus 100 detects the failure of the server A, (3) Server A is notified of the failure status of all terminals in use as a server. (4) When the terminal 200 receives the failure information of the ServerA from the central control device 100, it fetches the next Proxy Server ServerB information from the next proxy server list of the failed ServerA service server group 300, Is set as a new proxy server to perform failback. In this case, the UE first recognizes the failure before it receives the failure status of the SeverA from the central control unit 100. In this case, At this time, the UE can perform a fail-over first for fast failback and change (replace) the proxy server to the ServerB. After that, if the Server A failure in the step (3) is received from the central control device 100 Since the server A is not in use, it transmits the proxy information currently being used to the central control apparatus 100 without any action, thereby notifying the central control apparatus 100 of the problem.

FIG. 6 shows a processing procedure of a service system control method 2000 for providing policy application, load balancing, and failure recovery according to another embodiment of the present invention. The control method is a control method of a service system that provides policy application, load balancing, and failure recovery, including a central control device, one or more terminals, and a plurality of internal servers, A registration step of registering a policy server and a service server group capable of receiving load balancing and failover services; A failure and load detection step of periodically collecting the degree of load and failure of each internal server of the registered service server group; A terminal control rule setting step of transmitting a terminal control rule to a terminal upon start of a terminal, a policy modification, or a failure, so that a packet of the terminal can be processed as controlled by the central controller; A packet generation and delivery step of processing the packet by the terminal itself according to a set terminal control rule as controlled by the central control device; A failure detection and failure notification step of notifying whether or not a terminal failure occurs so that all the terminals connected to the failure server can recognize the terminal failure, or detecting the failure of the terminal itself; And completing the failure recovery by replacing the failed proxy server setting with the backup proxy server setting by the terminal itself using the next proxy server list; As shown in FIG. The control method of the service system providing the policy application, the load balancing, and the failure recovery can be provided to the external system as one service.

Referring to FIG. 6, a processing procedure of a service system control method 2000 for providing load balancing and failover according to another embodiment of the present invention will be described in more detail below. First, as a terminal and policy registration step (S100), a service policy for each terminal is registered in the central control unit 100 to register which packet of a terminal is to be serviced by the service server group 300, Registering whether load balancing and failure recovery functions are provided or not is performed to register a service server group to which load balancing and failure recovery functions are to be provided.

Next, in the failure and load detection step (S200), collecting the degree of load and failure of each server of the service server group that is provided with the load balancing and the failure recovery function, periodically collects information.

Next, a terminal control rule setting step (S300) is executed when a terminal is newly started or terminal control rule resetting of the terminal is required. The terminal control rule is transmitted to the terminal, Allows you to control server farm servers. In this step, the central control unit calculates the terminal control rules for each terminal and transmits the calculated terminal control rules to the respective terminals so that each terminal processes the packets as controlled by the central control apparatus, It is possible to apply a differentiated policy to each terminal and packet and to provide load balancing and failure recovery for a service server group.

Next, a packet generation and transmission step (S400) is performed in a normal packet communication step of the terminal. The terminal completing the terminal control rule setup is subjected to this step when proceeding with packet communication. In accordance with the terminal control rule, So that packets can go through a server of a specific service server group. Each of the terminals receives packets from the central control device and processes the packets generated by the terminal internal application according to a predetermined terminal control rule so that the packets are processed as controlled by the central control device.

Next, a failure detection and failure notification step (S500) is performed when a failure occurs in the server of the service server group that is being serviced. The failure is detected by detecting the failure, and when the central control apparatus 100 detects failure Or notifies the related terminal whether the failure has occurred, or the terminal itself detects the failure and ends at the time when the terminal recognizes the occurrence of the failure. The detection of the failure by the central control unit 100 may be performed by providing an interface for transmitting information such as RestAPI, Netconf, or a newly created diagnostic protocol. However, the use of the program or protocol is not limited to this. The invention is not specific to the invention, and various programs and protocols can be used to detect load and failure.

In step S600, when the terminal detects a failure of a proxy server connected to the terminal, the terminal changes the failed server to the next proxy server established through the terminal control rule, And performs fault recovery immediately. After the failure recovery, the information about the newly-replaced proxy server is transmitted to the central control apparatus 100, and the step is terminated.

FIG. 7 shows a processing procedure of a service system control method 3000 for providing policy application, load balancing, and failure recovery according to another embodiment of the present invention. Referring to FIG. 7, a processing procedure of a service system control method 3000 for providing load balancing and failover according to another embodiment of the present invention will be described below.

First, as a policy application step (S3100), a control method of a service system providing load balancing and failure recovery, including a central control device, one or more terminals, and a service server group composed of a plurality of internal servers, So that a differentiated policy can be applied to each packet via a different proxy server for each packet generated by the terminal.

Next, in the load balancing step S3200, the central control unit adjusts the proxy server IP of the terminal control rule to be downloaded to the terminal, and performs a load balancing function by connecting the proxy server IP to different proxy servers for each terminal.

Next, as a failure recovery performing step (S3300), the next proxy server list is received through the terminal control rule, and when a failure occurs in the currently connected proxy server, reconnection to the next proxy server is performed, Function.

In the processing procedure of the service system control method 3000 for providing policy application, load balancing, and failure recovery according to yet another embodiment of the present invention shown in FIG. 7, And the failure, the terminal control rule based on the preset terminal policy and the service server group policy, and transmits the calculated terminal control rule to the terminal, thereby applying differentiated service to each packet generated by the terminal Provides the load balancing between internal servers of the service server group and the failure recovery function of the internal server by providing a different proxy server for each terminal.

The at least one terminal sets a terminal proxy selector, a proxy client, or a next proxy server list according to a terminal control rule received from the central control device, and the packets generated by the terminal are transmitted to different proxy servers, Service is applied and the central control unit controls the proxy server set up for each terminal to provide load balancing and the proxy server is provided with a list of the next proxy server preset in the terminal so that the proxy server can be restored immediately when a failure occurs do. The proxy selector identifies a packet on the basis of a part of identifiers set in the packet identification information that can be used for packet identification such as L2 to L7 packet information and a terminal internal application identifier and then transmits the proxy client . The proxy client is connected to the proxy server supported by the server of the service server group among the available proxy client types, and transmits the packet to the corresponding proxy client using the proxy client type having the necessary function to allow the packet to be received via the proxy server The packet is set and transmitted so that the packet can pass through the proxy server connected to the proxy client. The next proxy server list is a list for storing information of backup proxy servers to be newly used when a failure occurs in a proxy server set for each proxy client. The list is stored for each proxy client, and when a failure occurs in a proxy server of a specific proxy client, The proxy server information of the proxy client is replaced with the backup proxy server immediately without inquiring to the device to complete the failure recovery and the server of the next proxy server list descending for each terminal is made different so that the failure recovery is performed simultaneously with load balancing, You can set up a serving server as a proxy server for backup so that failover can be provided without a standby server for backup.

The service server group is a set of servers that provide services that are differentially provided for each terminal and each packet. Each service server group provides specific services specific to each policy, so that the terminal can forward So that different services can be applied to each packet.

FIG. 8 shows an example of the configuration of a service system according to the present invention. Referring to FIG. 8, an MPTCP gateway service capable of providing MPTCP performance to a final server 400 that can not use MPTCP (Multipath TCP) , And failure recovery.

MPTCP (Multipath TCP) is a technology that establishes a TCP connection by bundling each path if there are multiple paths between the starting point and the destination, and improves the packet transmission speed between the destination and the starting point by enabling the merging transmission of the packets. Currently, mobile devices using the mobile communication network can be connected to two networks, such as LTE / 3G and WiFi, so that they can perform merge transmission and speed up through merge transmission.

As shown in FIG. 8, in the current Internet network, most servers do not support MPTCP, so even if the terminal has a function of using MPTCP, it is not receiving the benefits. Therefore, the MPTCP gateway (MPTCP gateway) applies MPTCP to the backbone network in order to provide an improved speed through the merge transmission for high-speed traffic such as video traffic, and applies the MPTCP to the backbone to the final server 400, To provide a solution that provides increased speed due to merge transmission. In this case, the MPTCP Gateway can also benefit from the system of the present invention as one service server group.

As shown in FIG. 8, the terminal 200 first receives a terminal control rule from the central control device 100. FIG. After that, the terminal sends the packet generated by the application to the proxy selector and receives differentiated services for each packet. At this time, the packet receiving the MPTCP service is transmitted to the proxy and set to pass through the MPTCP Gateway 600, and then the packet is released using both the LTE / 3G network and the WiFi network through the MPTCP daemon. It can be confirmed that it is released. In addition, since the MPTCP Gateway is subscribed to the load balancing service, it is possible to confirm that the packet of the terminal 1 and the packet of the terminal 2 are distributed to Server_A and Server_B, respectively.

Meanwhile, a service system control method for providing policy application, load balancing, and failure recovery according to an embodiment of the present invention may be implemented in a form of a program command that can be executed through a variety of electronic information processing means, Lt; / RTI > The storage medium may include program instructions, data files, data structures, and the like, alone or in combination.

Program instructions to be recorded on the storage medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of software. Examples of storage media include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, magneto-optical media and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. The above-mentioned medium may also be a transmission medium such as a light or metal wire, wave guide, etc., including a carrier wave for transmitting a signal designating a program command, a data structure and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as devices for processing information electronically using an interpreter or the like, for example, a high-level language code that can be executed by a computer.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the appended claims. It will be understood that the invention may be varied and varied without departing from the scope of the invention.

100: central controller 101: policy handler
102: Policy Detector 103: Policy DB
104: terminal controller 105: policy generator
200: terminal 201: terminal control module
202: Terminal control rule
211: Agent 212: Proxy selector
213: proxy client 214: server list
300: service server group
400: final server
500: Manager

Claims (23)

1. A service system for providing policy application, load balancing, and failover, including a central control device, a service server group composed of one or more terminals, and a plurality of internal servers,
A differentiated policy can be applied to each packet via a different proxy server selected from internal servers of the service server group for each packet generated by the terminal according to a terminal control rule preset by the central control device, The apparatus adjusts the proxy server IP of the terminal control rule to be downloaded to the terminal so that the proxy server can be connected to different proxy servers for each terminal so that the load balancing function can be performed. And a proxy server connected to the current terminal, and when the failure occurs in the current proxy server, the proxy server is reconnected to the next proxy server to provide a failure recovery function to the proxy server. , And a service system that provides failover.
2. The apparatus of claim 1, wherein the central control device
A terminal control rule is calculated based on a degree of load and failure of each internal server of the service server group, a preset policy for each terminal and a policy of a service server group, and the calculated terminal control rule is transmitted to the terminal, A load balancing function for providing load balancing between internal servers of the service server group and a failure recovery function of the internal server by providing differentiated application of services to packets and differentiating proxy servers to be connected to each terminal, And a service system that provides failover.
The method of claim 1, wherein the one or more terminals
A proxy client, or a proxy server according to the terminal control rules received from the central control unit, and the packets generated by the terminal are transmitted to different proxy servers so that differentiated services are applied to the packets The central control device can control the proxy server set for each terminal to provide load balancing, and the proxy server can be restored immediately when a failure occurs in the proxy server through a list of the next proxy servers preset in the terminal A service system that provides policy enforcement, load balancing, and failover.
4. The apparatus of claim 3, wherein the proxy selector
L2, and L7, and packet identification information that can be used for packet identification such as a terminal internal application identifier, identifies a packet based on a part of the identifiers set, and transmits the identified packet to a proxy client To a service system that provides policy application, load balancing, and failover.
4. The method of claim 3, wherein the proxy client
A proxy server which is connected to a proxy server of a service server group and has a function necessary to allow a packet to be received via a proxy server so that a packet forwarded to the proxy client can be transmitted through a proxy server And distributing the service to the service system.
4. The method according to claim 3, wherein the next proxy server list
This is a list that keeps the information of the proxy server for backup when a failure occurs in the proxy server set for each proxy client. One proxy client is stored for each proxy client, so that when a failure occurs in the proxy server of the specific proxy client, The proxy server information of the corresponding proxy client is replaced with the proxy server so that the failure recovery is completed and the server of the next proxy server list descending for each terminal is made different so that the failure recovery is performed simultaneously with the load balancing. Wherein the server is capable of providing failover without using a standby server for backup by providing a policy application, a load balancing, and a failure recovery.
The method according to claim 1, wherein the service server group
A set of servers providing services that are provided differentially for each terminal and each packet. Each service server group provides specific services specific to each policy and sets the terminal to pass through another specific service server group for each packet, And a service system that provides policy application, load balancing, and failure recovery.
A control method of a service system that provides load balancing and failback,
A failure and load detection step in which the central control device periodically collects the degree of load and failure of each internal server of the service server group;
A packet generation and delivery step in which a packet generated in the terminal is transmitted via an internal server of the service server group according to a terminal control rule set in the terminal by the central control device;
When the internal server of the service server group fails, the central control unit informs the terminal which is in connection with the failed server that the corresponding server has failed, or detects the failure of the internal server which is being operated by the terminal itself. step; And
A failure recovery step of completing a failure recovery by replacing a failed proxy server setting with a backup proxy server setting by using a next proxy server list established according to a terminal control rule by the terminal; Wherein the policy application, the load balancing, and the failback are provided.
9. The method according to claim 8,
And a registration step of registering a service server group to receive a load balancing and a failure recovery service and internal servers of the server group, And a method of controlling a service system that provides failover.
9. The method of claim 8, wherein the failure and load sensing step
The central control device collects load information and fault information from a service server group periodically applying the load balancing and / or failure recovery service through a predetermined protocol or program from the service server group. And a method of controlling a service system that provides failover.
9. The method according to claim 8,
The central control unit calculates the terminal control rules for each terminal and transmits the calculated terminal control rules to the terminals so that each terminal processes the packets according to the setting of the terminal control rules, And a terminal control rule setting step of providing load balancing and failure recovery for the service server group. The method of controlling a service system providing policy application, load balancing, and failure recovery.
9. The method of claim 8,
Wherein each of the terminals receives packets from the central control device and processes the packets generated by the terminal internal application according to a predetermined terminal control rule, so as to provide a policy application, a load balancing, and a failure recovery.
9. The method of claim 8, wherein the fault detection and fault notification step
The central control unit detects the failure, and notifies the terminals connected to the failed server of the failure, or each terminal detects the failure by itself, thereby allowing the terminal to recognize the failure. , And a control method of a service system that provides failover.
9. The method of claim 8,
And the failure proxy server is immediately replaced with the proxy server for backup in the next proxy server list preset in receipt of the terminal control rule immediately after the terminal recognizes the failure, And a method of controlling a service system that provides failover.
1. A control method of a service system that provides policy application, load balancing, and failback, including a central control device, a service server group composed of one or more terminals, and a plurality of internal servers,
Applying a differentiated policy to each packet via a proxy server selected among internal servers of the service server group for each packet generated by the terminal according to a terminal control rule preset by the central control device;
A load balancing step of controlling a proxy server IP of a terminal control rule provided to the terminal by the central control device so that the central control device is connected to a different proxy server for each terminal to perform a load balancing function; And
The terminal receives the list of the next proxy server including the information about the next proxy server through the terminal control rule and performs reconnection to the next proxy server even when a failure occurs in the proxy server connected to the current terminal, And a failure recovery performing step of providing a failure recovery function that provides a failure detection function and a failure recovery function.
16. The apparatus of claim 15, wherein the central control device
Service server group The terminal control rule is calculated based on the load degree and the failure degree of each internal server of the service server group, the preset terminal policy and the service server group policy, and the calculated terminal control rule is transmitted to the terminal. The load distribution between the internal servers of the service server group and the failure recovery function of the internal server are provided by providing differentiated application of services to each packet of the service server group and different proxy servers connecting to each terminal, A method of controlling a service system that provides failover.
16. The method of claim 15, wherein the one or more terminals
A proxy client, or a proxy server according to the terminal control rules received from the central control unit, and the packets generated by the terminal are transmitted to different proxy servers so that differentiated services are applied to the packets The central control device can control the proxy server set for each terminal to provide load balancing, and the proxy server can be restored immediately when a failure occurs in the proxy server through a list of the next proxy servers preset in the terminal A method of controlling a service system that provides policy enforcement, load balancing, and failover.
18. The apparatus of claim 17, wherein the proxy selector
L2, and L7, and packet identification information that can be used for packet identification such as a terminal internal application identifier, identifies a packet based on a part of the identifiers set, and transmits the identified packet to a proxy client To a service system that provides policy application, load balancing, and failover.
18. The method of claim 17, wherein the proxy client
A proxy server which is connected to a proxy server of a service server group and has a function necessary to allow a packet to be received via a proxy server so that a packet forwarded to the proxy client can be transmitted through a proxy server And distributing the load of the policy service to the service system.
18. The method of claim 17, wherein the next proxy server list
This is a list that keeps the information of the proxy server for backup when a failure occurs in the proxy server set for each proxy client. One proxy client is stored for each proxy client, so that when a failure occurs in the proxy server of the specific proxy client, The proxy server information of the corresponding proxy client is replaced with the proxy server so that the failure recovery is completed and the server of the next proxy server list descending for each terminal is made different so that the failure recovery is performed simultaneously with the load balancing. And a failover can be provided without a standby server for backup by allowing the server to be set up as a server, and a control method of a service system that provides policy application, load balancing, and failover.
16. The method according to claim 15, wherein the service server group
A set of servers providing services that are provided differentially for each terminal and each packet. Each service server group provides specific services specific to each policy and sets the terminal to pass through another specific service server group for each packet, Wherein the service application system is configured to allow different services to be applied to the service system.
A method for controlling a service system, the method comprising: applying a policy application, a load distribution, and a control method of a service system providing fault recovery according to any one of claims 8 to 21 to the external system as one service, A method of controlling a service system that provides recovery.
A computer-readable recording medium storing a program for executing the method of any one of claims 8 to 21.

Images