JP4970565B2 - Network routing method - Google Patents

Description

  The present invention relates to a network route selection method for selecting a route when a client device connects to a predetermined server device through a VPN (Virtual Private Network) in a network in which server devices existing at a plurality of bases are connected to each other.

In recent years, research and development on cloud computing technology has been promoted. The technology of the server environment that constitutes the cloud is advancing day by day, but there are still many issues to be solved regarding the client environment and network environment of the users who use the services provided by the cloud. Is left.

  For example, a company having branch offices or sales offices all over the country may have servers in a plurality of data centers in remote locations. As a remote access method for accessing a data center server from a branch office or sales office in such a corporate network, a network line such as a dedicated line or Internet VPN between the branch office or sales office and the data center, or a combination of these lines. Use to access and communicate. A dedicated line is a line connecting two specific points, and the line is managed by a telecommunications carrier, so it is very excellent in terms of communication quality and security.

On the other hand, since the Internet VPN uses an Internet network, it can communicate at a lower cost than a dedicated line . However, the Internet network has no bandwidth guarantee, and there is a problem that the communication speed is lowered by processing such as encapsulation and encryption for ensuring security.

  For example, if there is a data center in Kyushu, if users use Internet VPN from Tokyo to Kyushu, or if multiple users access the same site via Internet VPN at the same time, line delay will occur. appear.

  In order to solve such problems, it is possible to make the communication line as stable as possible by reporting information on the bandwidth and load status between routers and controlling the amount of transmitted packets according to the line condition. It is possible to introduce a dynamic VPN service that can be used. Patent Document 1 is cited as a technique for solving the decrease in communication speed.

JP 2005-223375 A

  However, in the dynamic VPN service, it is necessary to reconstruct the network in order to introduce it, so that it takes a period of time for design and construction, and a lot of costs are associated with the period. The technique of Patent Document 1 may be used in an environment where a plurality of VPN tunnels are formed through a plurality of providers to broadcast IP packets, but cannot be used in other cases.

  The present invention has been made in view of the above circumstances, and in a network in which server devices existing at a plurality of bases are connected to each other, when the client device is connected to the Internet VPN, it is possible to deal with a line congestion situation or a line failure. By selecting a compatible communication path, access via a stable communication path is possible, and there is no need to add equipment or rebuild the network environment from the existing network environment, which can be realized economically. An object of the present invention is to provide a network routing method.

  In the network route selection method according to an aspect of the present invention, a client device connects to a predetermined server device via a VPN server device via a VPN in a network in which server devices respectively present at a plurality of bases are connected to each other through a dedicated line. A network path selection method for selecting a path of a server, wherein when the client device is not connected to the VPN by a throughput measuring means, a VPN server device is connected to a server device existing at each of the plurality of bases. And connect the client device to measure the throughput value between the client device and each server device, or the client remotely connects to the measurement server via SSH and executes a command to measure the throughput from the measurement server to the client. And get the measurement result In addition, the client acquires the throughput value between the server devices at each site using SSH, saves the measured or acquired throughput value in a predetermined area of the storage medium, and disconnects the connected VPN connection after saving. The client device is based on the respective throughput values stored in the predetermined area of the storage medium by the route selection means, and is the most possible route that the client device can connect to the predetermined server device. A route having a high throughput is selected, and the address of the VPN server located in the selected route is set in the hosts file of the client device.

  The network route selection method according to another aspect of the present invention is the network route selection method according to claim 1, wherein the server devices existing at each of the plurality of bases measure the throughput value between the server devices by the throughput measurement unit. Then, the measured throughput value between the server devices is stored in a predetermined area of the storage medium.

  According to another aspect of the present invention, there is provided a network route selection program in which a client device is connected to a predetermined server device via a VPN server device via a VPN in a network in which server devices respectively existing at a plurality of bases are connected to each other through a dedicated line. A network route selection program for selecting a route when the client device is connected to a server device existing at each of the plurality of bases via a VPN server device when the client device is not connected to the VPN. Measure the throughput value between the device and each server device, or the client remotely connects to the measurement server via SSH, executes the command to measure the throughput from the measurement server to the client, and acquires the measurement result , The client uses SSH A throughput measurement function for acquiring a throughput value between server devices at bases, storing each measured or acquired throughput value in a predetermined area of the storage medium, and disconnecting a connected VPN connection after the storage; Based on the respective throughput values stored in the predetermined area, the VPN server that selects the path with the highest throughput among the paths that the client apparatus can connect to the predetermined server apparatus and is located on the selected path And a path selection function for setting the address in the hosts file of the client device.

  A client device according to another aspect of the present invention selects a path for connecting to a predetermined server device via a VPN server device via VPN in a network in which server devices existing at a plurality of bases are connected to each other by a dedicated line. When the client device is not VPN-connected, the client device is connected to a server device existing at each of the plurality of bases via a VPN server device, and the client device and each server Measure the throughput value with the device, or the client remotely connects to the measurement server with SSH, executes the command to measure the throughput from the measurement server to the client, acquires the measurement result, and the client uses the SSH Throughput value between server devices at each site The obtained throughput value is stored in a predetermined area of the storage medium, the throughput measurement means for disconnecting the connected VPN connection after the storage, and each of the throughput values stored in the predetermined area of the storage medium Based on the throughput value, the route with the highest throughput is selected from the routes that the client device can connect to the predetermined server device, and the address of the VPN server located in the selected route is the hosts file of the client device. And a route selection means for setting to (1).

  According to the present invention, in a network in which server devices existing at a plurality of bases are connected to each other, when a client device is connected to the Internet VPN, a communication path corresponding to a line congestion state or a line failure is selected. Provides a network route selection method that enables access via a stable communication route and that can be realized economically without the need to add devices or reconstruct the network environment from the existing network environment. be able to.

The figure which shows typically the 1st communication form of the network system to which the network path | route selection method which concerns on one Embodiment of this invention is applied. The figure which shows the 2nd communication form of the network system typically. FIG. 3 is a block diagram showing an example of a functional configuration of a client 3. The block diagram which shows an example of a function structure of the servers 11 and 21. FIG. The figure which shows an example of the content of the setting file with which the client 3 is provided. The figure which shows an example of the content of the throughput measurement result log file (measurement result file) with which the servers 11, 21 and the client 3 are provided. The figure which shows an example of the content of the setting file with which the servers 11 and 21 are provided. The flowchart which shows the first half part of the operation | movement by the application 30 with which the client 3 is provided. The flowchart which shows the second half part of the operation | movement by the application 30 with which the client 3 is provided. 10 is a flowchart showing the operation when the priority measurement value is “2” in the operation of FIG. 9. 10 is a flowchart showing the first half of the operation when the priority measurement value in the operation of FIG. 9 is “3”. 10 is a flowchart showing the second half of the operation when the priority measurement value is “3” in the operation of FIG. 9. The flowchart which shows the operation | movement by the application 10 with which the servers 11 and 21 are provided.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram schematically showing a first communication form of a network system to which a network route selection method according to an embodiment of the present invention is applied. FIG. 2 is a diagram schematically showing a second communication form of the network system.

  In this embodiment, for convenience, the server device is referred to as a “server” and the client device is referred to as a “client”.

  As shown in FIGS. 1 and 2, for example, one or a plurality of servers exist in each of a plurality of bases in units of data centers 100, 200,. The data centers are connected to each other with a dedicated line. A dedicated line is a dedicated line for data communication between two specific points, and in this embodiment, VPN servers (routers) at each base are connected to each other by this line. The VPN server has a server / router function for performing encapsulation and encryption for VPN communication. When starting the VPN connection, the client designates the host name of the VPN server.

  The server group shown in FIGS. 1 and 2 constitutes a cloud environment as a whole, and can provide various services to individual clients through the Internet VPN.

  In the example of FIGS. 1 and 2, the data center 100 includes a VPN server (router) 1, and a server 11 and a server 12 connected to the VPN server (router) 1. At least one of the server 11 and the server 12, for example, the server 11, resides in an application 10 that operates using information of a setting file to be described later. The data center 200 includes a VPN server (router) 2, a server 21 and a server 22 connected to the VPN server (router) 2. At least one of the server 21 and the server 22, for example, the server 21, resides in the same thing as the application 10. The VPN server (router) 1 and the VPN server (router) 2 are connected by a dedicated line, and each server in the data center 100 and each server in the data center 200 can communicate with each other via the dedicated line.

  The client 3 is a terminal used by the user. The client 3 can communicate by connecting to a predetermined server at each site via the Internet VPN. The Internet VPN is a virtual private network constructed via the Internet. By using the Internet VPN, it is possible to operate in the same manner as connecting remote networks with a LAN while maintaining confidentiality. Further, the client 3 is provided with an application 30 that operates using information of a setting file described later. Further, the client 3 is provided with a “hosts file” that is a network system file for host name resolution, and a host name and an IP address of a connection destination can be set through an input device or a display device.

  Here, a general procedure for forming the first communication form shown in FIG. 1 will be described.

For example, the communication unit receives the host name of the VPN connection destination which is an arbitrary character string input by the user from the VPN connection screen through the display device. Consider the case where the host name is set in the hosts file. Here, it is assumed that the host name of the VPN connection destination points to the data center 100. The client 3 starts an Internet VPN connection to the IP address corresponding to the host name via the communication unit . Here, it is assumed that the IP address corresponding to the host name is the IP address of the VPN server 1. Thereby, the VPN connection between the client 3 and the data center 100 (VPN server 1) is established. Next, the communication unit receives the IP address of the server to be connected input by the user through the connection screen. Here, it is assumed that the IP address is the IP address of the server 11. As a result, the VPN server 1 confirms the connection information of the designated IP address and starts communication between the client 3 and the server 11.

  Next, a general procedure for forming the second communication form shown in FIG. 2 will be described.

Again, when the communication unit receives the host name of the VPN connection destination input by the user through the connection screen of the client 3, the host name is set in the hosts file. Here, it is assumed that the host name of the VPN connection destination points to the data center 200. The client 3 starts Internet VPN connection to the IP address corresponding to the host name via the communication unit . Here, it is assumed that the IP address corresponding to the host name is the IP address of the VPN server 2. Thereby, the VPN connection between the client 3 and the data center 200 (VPN server 2) is established. Next, the IP address of the server to be connected input by the user is received by the communication unit through the connection screen of the client 3. Here, it is assumed that the IP address is the IP address of the server 11. As a result, the VPN server 2 confirms the connection information of the designated IP address and starts communication between the client 3 and the server 11.

  Hereinafter, the functional configurations of the client 3 and the servers 11 and 21 will be described with reference to FIGS. 3 and 4 with reference to FIGS. 1 and 2.

  FIG. 3 is a block diagram illustrating an example of a functional configuration of the client 3 illustrated in FIGS. 1 and 2.

  The client 3 includes an application 30 that resides under an operating system (OS). The application 30 uses the communication unit 40 for communication with an external device, and storage units (storage media or storage devices) 41 to 44 for storing and reading information such as various files and lists. Is used. In the example of FIG. 3, the storage units 41 to 44 are arranged in different areas, but of course, all or some of them may be arranged together in one storage unit. The application 30 includes a throughput measurement unit 52 and a route selection unit 51 as functional modules.

  The path selection unit 51 has a function of instructing a throughput measurement by transmitting a measurement server IP address, which will be described later, to the throughput measurement unit 52 based on the contents of the setting file stored in the storage unit 41.

When the throughput measurement unit 52 receives a request from the route selection unit 51 and confirms that the client 3 does not have a VPN connection for service use , the route measurement unit 51 stores the setting file stored in the storage unit 41. Based on the described IP address, a VPN connection is made via a VPN server to measurement servers existing at a plurality of bases through the communication unit 40, and an uplink throughput value between the client 3 and each measurement server is measured. . Further, the client remotely connects to the measurement server via SSH, executes a command for measuring the throughput from the measurement server to the client, and acquires the measurement result. This throughput information is a downstream throughput value between the client and the server. Throughput measurement here performs the transmission of a predetermined amount of data to IP add-less servers to be measured, to measure taken to transmission time, a measurement of the transmission amount per unit time, commonly used It can be measured with any conventional throughput measuring machine.

  The client then uses SSH to obtain the throughput value between the measurement servers at each site (specifically, obtain the latest one-line throughput measurement value from the throughput measurement result log file of the measurement server at each site) And) each measured or acquired throughput value is stored in the storage unit 42 as a throughput measurement result log file (hereinafter sometimes referred to as a “measurement result file”), and the VPN connection being connected is disconnected after the saving. It is a function. SSH is a program for executing a command from, for example, a remote device or moving a file to another device using encryption or authentication technology.

  Further, the route selection unit 51 creates a route list on the storage unit 43 based on each throughput value stored in the storage unit 42, and also creates a calculation list as necessary. Using the list, the route with the highest throughput is selected from the routes that the client 3 can connect to a predetermined server, and the address of the VPN server located on the selected route is stored in the storage unit 44 of the client 3. Has a function of setting the hosts file stored in the file.

  In addition, the presence / absence of VPN connection for client service use is monitored.

  Hereinafter, the various elements shown in FIG. 3 will be described in more detail.

  The setting file stored in the storage unit 41 includes information necessary for the throughput measurement unit 52 to measure or acquire a throughput value between servers in each data center and to acquire a throughput value between servers. Have. As shown in FIG. 5, the setting file includes (1) operation interval time, (2) priority measurement value, (3) host name, (4) VPN server IP address (connection base), and (5) measurement server IP. Address (connection base), (6) VPN server IP address (route base), and (7) Measurement server IP address (route base).

  The “connection site” refers to a site where the connection server is located. Here, “connection server” refers to a server to which a client connects. On the other hand, the “route base” refers to a base through which a connection base is connected.

  (1) The operation interval time is an interval time for performing the throughput measurement. Specifically, it is an interval time for performing a series of operations from connection with each server to throughput measurement and rewriting of the hosts file. For example, assuming that the operation interval time is 5 minutes and there are two measurement servers, server A and server B, if server A is measured at AM 10:00 and all measurement operations take, for example, 1 minute, AM 10:01 Server B is measured, and then measurement of server A is started at AM 10:05 and server B is measured at AM 10:06.

  (2) The priority measurement value specifies which information of the average of the uplink and downlink of the throughput value of the selected route or the average of both is used at the time of route selection. In this embodiment, “1” indicates that priority is given to uplink, “2” indicates that priority is given to downlink, and “3” indicates that priority is given to the average of uplink and downlink.

  (3) The host name is the host name of the base where the connection server exists, and is the target host name for which the selected route is to be rewritten. This host name is used when making a VPN connection.

  (4) The VPN server IP address (connection base) is the IP address of the VPN server at the base where the connection server exists.

  (5) The measurement server IP address (connection base) is the IP address of the server used for the throughput measurement of the base where the connection server exists.

  (6) VPN server IP address (route base) is the IP address of the VPN server existing at the route base.

  (7) The measurement server IP address (route base) is an IP address of a server used for throughput measurement at the route base.

  The throughput measurement result log file (measurement result file) stored in the storage unit 42 has a throughput value between the servers measured or acquired by the throughput measurement unit 52. As shown in FIG. 6, this measurement result file has information such as (1) measurement start time and (2) throughput measurement value.

  (1) The measurement start time is the time when the measurement is started.

  (2) The measured throughput value is a measured throughput value (Mbps).

  The route list and the calculation list stored in the storage unit 43 have information necessary for the route selection unit 51 to select a route. In the route list, a list in which VPN server IP addresses and throughput measurement values of all routes that can be connected to the connection server are associated is described. In the calculation list, the average value of the uplink throughput measurement value of each route, the downlink throughput measurement value, and the uplink and downlink throughput values of the route connectable to each connection server are described.

  The hosts file stored in the storage unit 44 has, in addition to the host name, the address of the VPN server to be connected to the VPN corresponding to the host name. In the hosts file, the address of the VPN server located on the route selected by the route selection unit 51 is set.

  FIG. 4 is a block diagram illustrating an example of a functional configuration of the servers 11 and 21 illustrated in FIGS. 1 and 2.

  Each of the servers 11 and 21 includes an application 10 that resides under an operating system (OS). The application 10 uses a communication unit 60 for communication with an external device, and uses storage units (storage media, storage devices) 61 and 62 for storing and reading various files. In the example of FIG. 4, the storage units 61 and 62 are arranged in separate areas, but of course, all or some of them may be arranged together in one storage unit. The application 10 includes a measurement control unit 63 and a throughput measurement unit 64 as functional modules.

  The measurement control unit 63 has a function of sending a measurement server IP address or the like to the throughput measurement unit 64 and requesting a throughput measurement based on the contents of the setting file stored in the storage unit 61. The throughput measuring unit 64 measures the throughput value with the server at each site described in the setting file in accordance with the request from the measurement control unit 63, and stores the measured throughput value in the storage unit 62 as a measurement result file. .

  Hereinafter, the various elements shown in FIG. 4 will be described in more detail.

  The setting file stored in the storage unit 61 has information necessary for the throughput measurement unit 64 to measure a throughput value with a server at another site. As shown in FIG. 7, the setting file has information such as (1) operation interval time and (2) measurement server IP address.

  (1) The operation interval time is an interval time for performing the throughput measurement, and the operation is the same as that of the client application.

  (2) The measurement server IP address is an IP address of a server used for throughput measurement at another base.

The throughput measurement result log file (measurement result file) stored in the storage unit 62 has the throughput value measured by the throughput measurement unit 64. The file name is specified by a combination of “measurement source IP address”, “measurement destination server IP address”, and “YYYYMMDD (date of the day)”. As shown in FIG. 6, this measurement result file has information such as (1) measurement start time and (2) throughput measurement value.

  (1) The measurement start time is the time when the measurement is started.

  (2) The measured throughput value is a measured throughput value (Mbps).

(Outline of operation of application 30 on client 3 side)
The outline of the operation of the application 30 on the client 3 side is shown below.

1) Processing at start-up-The application 30 is activated when the OS of the client 3 is activated.

  -After activation, the route selection unit 52 reads the setting file.

  The route selection unit 52 checks whether or not the client 3 is performing VPN connection for service use.

  The throughput measuring unit 52 establishes VPN connection to the measurement server at each site read from the setting file via the communication unit 60 when the service use VPN is not connected, and starts processing such as VPN throughput measurement.

  • If the client is connected to the service using VPN, the process ends without performing any processing.

2) Throughput measurement processing, information collection / accumulation processing-Establish VPN connection with the VPN server at each site described in the configuration file and the communication unit 40, measure the uplink and downlink throughput, and measure the uplink and downlink throughput measurement values. Add to the throughput measurement result log file (measurement source IP address_measurement destination server IP address_YYYYMMDD (date of the day) .log) in the storage unit 42 or newly save it.

  For uplink throughput measurement, the throughput from the client 3 to the measurement server is measured.

  For downlink throughput measurement, the client 3 remotely connects to the measurement server via SSH, executes a command for measuring the throughput from the measurement server to the client 3, and acquires the measurement result.

  The client 3 uses SSH to obtain only the latest line of the throughput measurement result log file between the servers in the measurement server at each site.

  Disconnect the VPN connection.

  The above processing is repeated for the number of server IP addresses of each base set in the setting file.

3) Route creation processing, route setting processing-The route selection unit 51 refers to the priority measurement value in the setting file and performs route selection processing.

  ・ Also, before selecting a route, it is possible to connect to the VPN server IP address and throughput measurement value of all routes, and the uplink and downlink throughput measurement values of each route and each connection server. A calculation list for storing the average value of the throughput values of the uplink and downlink of the route is generated in the storage unit 43 on the memory.

  ・ The route selection method is as follows.

3-1) When the uplink throughput measurement value is prioritized-From the throughput measurement result log file of the client 3, the uplink throughput measurement value from the client 3 to the measurement server (connection base) is read, and the VPN server IP address of the configuration file Together with the contents of (connection base), it is stored in the route list of the storage unit 43 on the memory.

  ・ Read the IP address list of the measurement server (route base) from the configuration file and perform the following process.

  -From the throughput measurement result log file of client 3, read the uplink throughput measurement value from client 3 to the measurement server (route base) and the uplink throughput measurement value from the measurement server (route point) to the measurement server (connection site), Comparing the throughput measurement values from the client 3 to the measurement server (route base) and the measurement server (route base) to the measurement server (connection site), the uplink throughput measurement value of the lower throughput measurement value and the measurement server (route base) Are stored in the route list of the storage unit 43 on the memory.

  Select the route with the highest throughput measurement value stored in the route list.

3-2) When the downlink throughput measurement value is prioritized • The downstream throughput measurement value from the client 3 to the connection site is read from the throughput measurement result log file of the client 3, and the VPN server IP address (connection site) in the configuration file Are stored in the path list of the storage unit 43 on the memory.

  Read the list of measurement server IP addresses from the configuration file, and perform the following processing for each measurement server.

-From the throughput measurement result log file of client 3, read the downstream throughput measurement value from client 3 to the measurement server (route base) and the downlink throughput measurement value from measurement server (route base) to measurement server (connection base). The throughput measurement values from the client 3 to the measurement server (route base) and the measurement server (route base) to the measurement server (connection site) are compared, and the downstream throughput measurement value of the lower throughput measurement value is compared to the measurement server (route base). ) The VPN server IP address (route base) of the same base is stored in the route list of the storage unit 43 on the memory.

  Select the route with the highest throughput measurement value stored in the route list.

3-3) When the uplink and downlink averages are prioritized-From the client 3 throughput measurement result log file, the uplink and downlink throughput measurement values from the client 3 to the connection base are read, and the VPN server IP address ( Together with the content of the (connection base), it is stored in the calculation list of the storage unit 43 on the memory.

  Read the list of measurement server IP addresses from the configuration file, and perform the following processing for each measurement server.

  -Based on the measurement results of client 3, the uplink and downlink throughput measurement values from client 3 to the measurement server (route base) and the uplink and downlink throughput measurement values from the measurement server (route base) to the measurement server (connection site) The data is read and compared separately for uplink and downlink, and the lower throughput measurement value is stored in the calculation list of the storage unit 43 on the memory as the uplink and downlink throughput measurement value.

  Calculate the average value of all uplink and downlink throughput measurements from the calculation list.

  Calculate the ratio of the throughput measurement value to the average value for each uplink and downlink separately for each route, calculate the ratio of the throughput measurement value to the route list, and store the multiplied value and the VPN server IP address in the route list To do.

  -Select the route with the highest throughput ratio stored in the route list.

4) Route information update processing When the IP address corresponding to the host name in the host file of the client is different from the IP address of the VPN server between the client and VPN server of the route selected above, the IP address is changed.

  If they are the same, do nothing.

5) Processing at the end ・ When the client 3 starts the VPN connection for using the service, a series of processing performed by the application 30 of the client 3 ends.

  When the VPN connection using the client 3 service is terminated, a series of processing by the application 30 is performed only once.

(Outline of operation of application 10 on server 11 and 21 side)
The outline of the operation of the application 10 on the server 11 and 21 side will be described below.

1) Processing at start-up-The application 10 is activated when the OS of the server is activated.

2) Throughput measurement processing and information collection / accumulation processing-Connection to the measurement server specified in the setting file is started by the communication unit 60, and throughput measurement is performed by the throughput measurement unit 64. At this time, the server measures the uplink throughput with the measurement destination server, and the throughput measurement value log file (measurement source IP address_measurement destination server IP address_YYYYMMDD (today) (Date) .log) Add or save a new one.

3) Processing at the end ・ Repeat the above processing for the number of server IP addresses of each base set in the setting file.

(Operation flow of the application 30 on the client 3 side)
Next, the operation of the application 30 on the client 3 side will be described with reference to the flowcharts of FIGS.

  For the sake of explanation, the path between the client 3 and the measurement server (connection base) is line A, the path between the client 3 and the measurement server (route base) is line B, and the path between the measurement server (route base) and the measurement server (connection base). Is line C.

  When the OS of the client 3 is activated, the application 30 is activated. The activated application 30 resides in the OS, and repeats a series of processes as described below at regular intervals set in the setting file.

  The application 30 determines whether or not the client 3 has made a VPN connection for service use through the communication unit 40 (step S11 in FIG. 8). If the service is being used for VPN connection, nothing is done and the process is terminated. On the other hand, if the VPN for service use is not being connected, processing for measuring the throughput or the like is started.

Next, in the application 30, the path selection unit 51 reads the IP address of the measurement server at each site and the IP address of the VPN server at each site from the setting file in the storage unit 41 (step S12), and the throughput measurement unit 52 Requests the Internet VPN connection to each VPN server from the client 3 via the communication unit 40, and starts the connection (step S13).

  Next, in the application 30, the route selection unit 51 reads the information of the priority measurement value from the setting file (step S14), and the priority measurement value is “1” (upstream) or “3” (upstream / downstream average). It is determined whether or not there is (step S15). If not, the process proceeds to step S20. On the other hand, if applicable, the application 30 uses the throughput measuring unit 52 to measure the throughput from the client 3 to all measurement servers existing at each site (step S16).

  Note that the measurement result of the path in which the error has occurred is set to 0 when an error occurs during the throughput measurement of the client 3 or when it is not possible to connect to the measurement server when measuring the throughput between servers.

  Next, the application 30 determines whether or not a throughput measurement result log file having the same name (measurement source IP address_measurement destination server IP address_YYYYMMDD (today's date) .log) exists in the storage unit 42 of the client 3 ( Step S17). If it does not exist, a new throughput measurement result log file is newly created, and the throughput measurement result is stored in the storage unit 42 (step S18). On the other hand, if it exists, the throughput measurement result is added to the throughput measurement result log file in the storage unit 42 of the client 3 and overwritten and saved (step S19).

Next, it is determined whether or not the priority measurement value is “2” ( downlink ) or “3” (uplink / downlink average) (step S20). If not, the process proceeds to step S25. On the other hand, if applicable, the application 30 remotely connects to the measurement server via SSH, executes a command for measuring the throughput from the measurement server to the client 3, and acquires the measurement result (step S21).

  Next, the application 30 determines whether a throughput measurement result log file with the same name exists in the storage unit 42 of the client 3 (step S22). If it does not exist, a new throughput measurement result log file is created and the throughput measurement result is stored (step S23). On the other hand, if it exists, the throughput measurement result is added to the throughput measurement result log file in the storage unit 42 of the client 3 and overwritten and saved (step S24).

  Next, the application 30 remotely connects to the measurement server from the client 3 by SSH, and acquires only the latest throughput measurement value (only one line) from the throughput measurement result log file held by the measurement server (step). S25) The VPN connection is disconnected (step S26), and it is determined whether there is a measurement server (routed base) that has not been measured (step S27). If there is a measurement server (route base) that has not been measured, the process returns to step S13. On the other hand, if there is no measurement server (route base) that has not been measured, the process proceeds to step S30 in FIG.

  The application 30 reads the information of the priority measurement value from the setting file by the route selection unit 51 (step S30 in FIG. 9), and determines whether the priority measurement value is “1” (up) (step S30). . If not, the process proceeds to step S31. On the other hand, if applicable, the application 30 uses the route selection unit 51 to read the uplink throughput measurement value of the line A from the throughput measurement result of the client 3 (step S32), and the throughput measurement value and the VPN server IP address (connection base). Is stored in the route list of the storage unit 43 on the memory (step S33).

Next, the application 30 reads one measurement server IP address (route base) from the setting file by the route selection unit 51 (step S34), and reads the uplink throughput measurement value of the line B from the throughput measurement result of the client 3 ( Step S35), the uplink throughput measurement value of the line C is read from the measurement result of the measurement server (route base) (step S36), and it is determined whether or not the uplink throughput measurement value of the line B is lower than the uplink throughput measurement value of the line C. Determination is made (step S37). When the throughput measurement value of the line B is low, the uplink throughput measurement value of the line B and the IP address (route base) of the VPN server are stored in the path list of the storage unit 43 on the memory (step S38). On the other hand, if the throughput measurement value of channel B is not a low, stores uplink throughput measurement value of channel C and the VPN server IP address (via bases) in the path list storage unit 43 on the memory (step S39).

  Next, the application 30 uses the route selection unit 51 to determine whether or not there is an IP address of an unread measurement server (route base) from the setting file (step S40). If it exists, the process returns to step S34. On the other hand, if it does not exist, the VPN server IP address of the route with the maximum throughput measurement value is read from the route list of the storage unit 43 on the memory (step S41), and the VPN server of the route with the maximum throughput measurement value is read. It is determined whether the IP address is equal to the IP address of the hosts file in the storage unit 44 on the memory (step S42). If they are equal, nothing is done and the series of processing ends. On the other hand, if they are not equal, the IP address of the hosts file is rewritten (step S43), and the series of processes is terminated.

  On the other hand, if the priority measurement value is “2” in step S31, the application 30 uses the route selection unit 51 to read the downlink throughput measurement value of the line A from the throughput measurement result of the client 3 (step in FIG. 10). S51), the throughput measurement value and the VPN server IP address (connection base) are stored in the path list of the storage unit 43 on the memory (step S52).

  Next, the application 30 reads one measurement server IP address (route base) from the setting file by the route selection unit 51 (step S53), and reads the downlink throughput measurement value of the line B from the throughput measurement result of the client 3 ( In step S54), the downlink throughput measurement value of line C is read from the measurement result of the transit base (step S55), and it is determined whether the downlink throughput measurement value of line B is lower than the downlink throughput measurement value of line C (step S54). S56). When the throughput measurement value of the line B is low, the downlink throughput measurement value of the line B and the IP address (route base) of the VPN server are stored in the path list of the storage unit 43 on the memory (step S57). On the other hand, if the throughput measurement value of line B is not low, the downlink throughput measurement value of line C and the IP address (route base) of the VPN server are stored in the path list of the storage unit 43 on the memory (step S58).

  Next, the application 30 uses the path selection unit 51 to determine whether or not there is an unread measurement server IP address from the setting file (step S59). If it exists, the process returns to step S53. On the other hand, if not, the process proceeds to step S41 in FIG.

  On the other hand, when the priority measurement value is “3” instead of “2” in step S31 of FIG. 9, the application 30 uses the path selection unit 51 to determine the throughput above and below the line A from the throughput measurement result of the client 3. The measured values are read (step S61 in FIG. 11), and the upper and lower throughput measured values and the VPN server IP address (connection base) are stored in the calculation list of the storage unit 43 on the memory (step S62).

  Next, the application 30 reads one measurement server IP address (route base) from the setting file by the route selection unit 51 (step S63), and reads the uplink throughput measurement value of the line B from the throughput measurement result of the client 3 ( In step S64), the uplink throughput measurement value of line C is read (step S65), and it is determined whether the uplink throughput measurement value of line B is lower than the uplink throughput measurement value of line C (step S66). When the throughput measurement value of the line B is low, the uplink throughput measurement value of the line B and the IP address (route base) of the VPN server are stored in the calculation list of the storage unit 43 on the memory (step S67). On the other hand, when the throughput measurement value of the line B is not low, the uplink throughput measurement value of the line C and the IP address (route base) of the VPN server are stored in the calculation list of the storage unit 43 on the memory (step S68).

  Next, the application 30 reads the downlink throughput measurement value of the line B from the throughput measurement result of the client 3 by the route selection unit 51 (step S69), and reads the downlink throughput measurement value of the line C (step S70). It is determined whether or not the downlink throughput measurement value is lower than the downlink throughput measurement value of the line C (step S71). When the throughput measurement value of the line B is low, the downlink throughput measurement value of the line B and the IP address (route base) of the VPN server are stored in the calculation list of the storage unit 43 on the memory (step S72). On the other hand, if the throughput measurement value of the line B is not low, the downlink throughput measurement value of the line C and the IP address (route base) of the VPN server are stored in the calculation list of the storage unit 43 on the memory (step S73).

  Next, the application 30 determines whether or not an unselected server IP address exists in the measurement server IP address read from the setting file by the route selection unit 51 (step S74). If not, the process returns to step S63. On the other hand, if it exists, the process proceeds to step S76 in FIG.

  Next, the application 30 calculates the average of all the uplink throughput measurement values and all the downlink throughput measurement values from the calculation list in the storage unit 43 on the memory by the path selection unit 51 (step S76). The ratio of the uplink and downlink throughput measurement values to the line A with respect to the average throughput measurement value is calculated (step S77), and the calculated uplink ratio and downlink ratio are multiplied (step S78), and the result of the multiplication and the VPN server The IP address (connection base) is stored in the route list of the storage unit 43 on the memory (step S79).

  Next, the application 30 reads one measurement server IP address (route base) from the setting file by the route selection unit 51 (step S80), and the uplink and downlink throughput measurement values of the line B from the throughput measurement result of the client 3 Is read (step S81), and the uplink and downlink throughput measurement values of the line C are read from the measurement result of the measurement server (route base) (step S82).

  Next, the application 30 uses the path selection unit 51 to determine whether or not the uplink throughput measurement value of the line B is lower than the uplink throughput measurement value of the line C (step S83). When the throughput measurement value of the line B is low, the ratio of the uplink throughput measurement value of the line B to the average uplink throughput measurement value is calculated (step S84). On the other hand, if the throughput of the line B is not low, the ratio of the uplink throughput measurement value of the line C to the average uplink throughput measurement value is calculated (step S85).

  Next, the application 30 causes the path selection unit 51 to store the calculated uplink throughput ratio in the calculation list in the storage area 43 on the memory as an uplink throughput measurement value ratio (step S86).

  Next, the application 30 causes the route selection unit 51 to determine whether or not the downlink throughput measurement value of the line B is lower than the downlink throughput measurement value of the line C (step S87). When the throughput of line B is low, the ratio of the downlink throughput measurement value of line B to the average downlink throughput measurement value is calculated (step S88). On the other hand, if the throughput of line B is not low, the ratio of the downlink throughput measurement value of line C to the average downlink throughput measurement value is calculated (step S90).

  Next, the application 30 causes the path selection unit 51 to store the calculated downlink throughput ratio in the calculation list in the storage area 43 on the memory as a downlink throughput measurement value ratio (step S91).

  Next, the application 30 multiplies the uplink throughput measurement value ratio and the downlink throughput measurement value ratio by the route selection unit 51 (step S91), and the multiplied throughput measurement value ratio and the measurement server (route base) read in S81. ) Is stored in the route list of the storage unit 43 in the memory (step S92).

  Next, the application 30 uses the route selection unit 51 to determine whether or not there is an unread measurement server IP address from the setting file (step S93). If it exists, the process returns to step S53. On the other hand, if not, the process proceeds to step S41 in FIG.

(Operation flow of the application 10 on the server 11 and 21 side)
Next, the operation of the application 10 on the servers 11 and 21 side will be described with reference to the flowchart of FIG.

  When the OS of the servers 11 and 21 is activated, the application 10 is activated. The application 10 after activation resides in the OS and repeats a series of processes as described below at regular intervals set in the setting file.

  The application 10 reads the setting file from a predetermined storage medium or storage device by the path selection unit 63 (step S101), and measures the throughput between the measurement servers by the throughput measurement unit 64 (step S102).

  When an error is returned from the measurement destination server, the measurement result is set to 0, and the subsequent processing is continued.

  Next, the application 10 determines whether or not a throughput measurement result log file having the same name (measurement source IP address_measurement destination server IP address_YYYYMMDD (today's date) .log) exists in the storage unit 62 of the measurement source server. (Step S103). If not, a new throughput measurement result log file is newly created, and the throughput measurement result is stored (step S104). On the other hand, if it exists, the throughput measurement result is added to the throughput measurement result log file in the storage unit 62 of the measurement source server, and is overwritten and saved (step S105).

  Next, the application 10 determines whether there is a server at a base that has not been measured (step S106). If there is a server at a base that has not been measured, the process returns to step S102. On the other hand, if there is no server at the base that has not been measured, the series of processing ends.

  As described in detail above, according to the present embodiment, the data center of each base is connected to at least two bases by a high-quality line such as a dedicated line, and the client and the base are connected by the Internet VPN. In such an environment, the client owns the IP address of an accessible server and measures the throughput between servers, and the server measures and maintains the throughput of servers connected to each other. The client collects throughput information held by the server, selects the route with the highest throughput, and updates the address in its own hosts file to the selected address, so that the user can connect to the target server via the Internet VPN. To provide a program that enables optimal access or communication over the conventional VPN connection by accessing the target server through the base of the line with the highest throughput that is interconnected when attempting connection with it can.

  In addition, the program can only be installed on the client terminal or server machine, and at least the problem of communication degradation can be solved without restructuring the network or adding equipment, and can be accessed via the optimum route. Therefore, it is possible to avoid a failure at the time of a line failure.

  The various functions and processing procedures described in the above embodiments are stored as a computer program in a computer-readable storage medium or storage device (for example, a semiconductor memory, an optical disk, or a magnetic disk), and are stored as necessary. May be read and executed by a processor. Such a computer program can also be distributed by transmitting from one computer to another computer via a communication medium.

  The present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

  1, 2 ... VPN server, 3 ... client, 10, 30 ... application, 11, 12, 21, 22 ... server, 100, 200 ... data center, 40, 60 ... communication unit, 41-44 ... storage unit, 51 ... Route selection unit, 52... Throughput measurement unit, 61, 62... Storage unit, 63... Measurement control unit, 64.

Claims (4)

A network route selection method for selecting a route when a client device is connected to a predetermined server device via a VPN server via a VPN in a network in which server devices respectively existing at a plurality of bases are connected to each other through a dedicated line. ,
When the client device is not VPN-connected by the throughput measuring means, the client device is connected to a server device existing at each of the plurality of bases via a VPN server device, and the client device and each server device Measure the throughput value between the server and the client remotely connect to the measurement server via SSH, execute the command to measure the throughput from the measurement server to the client, acquire the measurement result, and the client uses SSH To obtain the throughput value between the server devices at each base, store the measured or acquired throughput values in a predetermined area of the storage medium, and disconnect the VPN connection being connected after saving,
Based on the respective throughput values stored in the predetermined area of the storage medium by the path selection unit, the client apparatus has the highest throughput among the paths that the client apparatus can connect to the predetermined server apparatus. A network route selection method comprising: selecting a high route and setting an address of a VPN server located on the selected route in a hosts file of the client device. The network routing method according to claim 1, wherein
The server devices existing at each of the plurality of bases measure throughput values between server devices by using a throughput measuring unit, and store the measured throughput values between server devices in a predetermined area of a storage medium. Network routing method. A network path selection program for selecting a path when a client apparatus connects to a predetermined server apparatus via a VPN server via a VPN in a network in which server apparatuses existing at a plurality of bases are connected to each other by a dedicated line. And
When the client device is not VPN-connected, connect to the server device existing at each of the plurality of bases via the VPN server device, and measure the throughput value between the client device and each server device, or The client remotely connects to the measurement server via SSH, executes a command to measure the throughput from the measurement server to the client, acquires the measurement result, and the client uses SSH to determine the throughput value between the server devices at each site. A throughput measurement function for storing each measured or acquired throughput value in a predetermined area of the storage medium, and disconnecting the connected VPN connection after saving,
Based on the respective throughput values stored in the predetermined area of the storage medium, the route with the highest throughput is selected from the routes that the client device can connect to the predetermined server device. A network routing program for causing a computer to realize a routing function for setting an address of a located VPN server in a hosts file of the client device. A client device capable of selecting a path for connecting to a predetermined server device via a VPN server device via a VPN in a network in which server devices respectively present at a plurality of bases are connected to each other by a dedicated line,
When the client device is not VPN-connected, connect to the server device existing at each of the plurality of bases via the VPN server device, and measure the throughput value between the client device and each server device, or The client remotely connects to the measurement server via SSH, executes a command for measuring the throughput from the measurement server to the client, acquires the measurement result, and the client uses the SSH to determine the throughput value between the server devices at each site. A throughput measuring means for storing the measured or acquired respective throughput values in a predetermined area of the storage medium, and disconnecting the connected VPN connection after the storage;
Based on the respective throughput values stored in the predetermined area of the storage medium, the route with the highest throughput is selected from the routes that the client device can connect to the predetermined server device. A client device comprising: path selection means for setting an address of a located VPN server in a hosts file of the client device.

Images