JP3472540B2 - Server selection device, server selection method, and recording medium recording server selection program - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a server selection method and device for selecting a server device to be accessed by a user who uses the service when a plurality of server devices on a communication network provide the same service. In particular, the present invention relates to a technique for ensuring that the selected server can reliably secure the resources necessary for providing the service and that the selected server is close to the user.

[0002]

2. Description of the Related Art With the explosive spread of the Internet in recent years, an increase in the load on a server device that processes a request from a client on the Internet has become a serious problem. In order to solve this, a server load balancing method is generally used in which a plurality of servers having the same function are prepared and processing requests from clients are evenly distributed to the plurality of servers. The conventional method related to server load distribution is explained in more detail. The simplest method is to prepare multiple servers that provide the same service,
URLs of those servers (Uniform
Resource Locator). When the processing speed is so slow that a user who accesses one of these servers feels uncomfortable, the user can enjoy a high-speed service by accessing another server.

However, according to this method, it is necessary to notify the user of a plurality of servers in advance, and depending on the means for notifying, it becomes difficult to notify the change, addition, or deletion of the server. HTML is the most common notification method
This is a method of listing links to multiple servers on the (HyperText Markup Language) page, but the link destinations listed on the HTML page must be changed every time a server is added, changed, or deleted. , Is inconvenient from the viewpoint of homepage maintenance. In addition, since this method only selects the server based on the user's judgment, it is impossible to evenly distribute the load on each server, and it is difficult to adjust the server load based on the intention of the service provider.

At present, a server load balancer for solving these problems is put into practical use. The server load balancer is generally a DNS (Domain Name System) server with its functions expanded. The server load balancer includes means for accepting an address resolution request for a certain URL from a client and selecting one server from a group of servers managed by the server load balancer by a random method or a round robin method. ,
A means for notifying the client of the IP address is provided. With this function, it is possible to distribute processing requests to the server based on the intention of the service provider rather than the user's intention.

According to the description of Nikkei Communication March 20, 2000, p.114-123 , the server selection method is not limited to the random number and round robin methods described above, but the round robin is weighted. Server selection based on various parameters such as method, method of assigning priority to servers, method of allocating server with least number of connections, method of allocating server with fastest response time, method of allocating server with lightest CPU load, etc. There is a server load balancer that performs As one of the server selection methods used for this server load balancer, etc., the paper “Locating Near Copies of Replicated Internet
net Servers ”(computer communication review vol.
25, no.4 p.288 (1995)), the distance between the server and the client (the number of hops) is obtained, and the method of selecting the server with the shortest distance as the server corresponding to the processing of the client is mentioned. The server selection device is provided with means for holding route information (router address) from the server selection device to each server to be controlled and acquiring route information from the server selection device to the client. An evaluation value that reflects the distance (hop count) between the server and the client is obtained from these pieces of information, and the server having the smallest evaluation value is selected. As a result, the user can access the closest server, and can suppress a decrease in response due to network congestion.

The above-mentioned method is a method mainly used in a server selection device for realizing load distribution processing for a current Web server, and access by the upper limit of the number of simultaneous accesses and the number of server licenses like the Web server is performed. It was a method that could effectively balance the load on servers with a relatively loose number limit. However, in the future, due to the performance of conference servers, moving image splitter servers, streaming servers, etc., which have not been fully considered as load balancing targets, conditions such as the number of licensed access and the transmission rate are restricted. You also have to consider relatively strict servers. When selecting a server for these servers, a server that can provide a service (for example, a server whose number of server accesses has not reached the upper limit determined for each server, or a transmission rate that allows the content transmission rate) Servers that have not reached the upper limit) must be selected.

At present, there are not many examples of a server selection device having such a function, but for example, Japanese Patent Laid-Open No. 11-4158.
As in the method described in Japanese Patent Publication No. 7, there are a plurality of video servers for sending moving images to clients, and the video management apparatus described in Japanese Patent Laid-Open No. 11-41587 described above delivers video titles from clients. There is an invention in which, when a request is received, a server with a light load is selected from a plurality of servers that can deliver the video title. This mechanism allows the client to use a video server with a light load and improve the delivery efficiency.

Further, in the server selection device itself, if the number of requests from the client temporarily increases, the server selection process becomes slow, and the response speed of the server selection process decreases. In order to solve these problems, at present, two or more server selection devices are duplicated and deployed at the same location, or, for example, the US-based CISCO load balancing device, Distribute Director, is also used.
Disabling which server group to process by combining the local load balancer Local Director
The Local Dire is determined by the tribute director and which server is selected from the servers in a certain area.
There is also a method of geographically distributing server selection devices by the method of determining by ctor.

[0009]

However, the server selection device described so far has the following problems. A conventional server load balancer selects a server from a group of servers registered in advance in the load balancer regardless of whether or not the selected server can provide the service requested by the user. The commonly used server selection method is based on various parameters such as random or round robin, weighted round robin, and server processing load (CPU load, response speed, etc.). It does not have a function to specify and specify an appropriate server from the servers whose remaining resource amount exceeds the required resource amount.

Therefore, in practice, a server that cannot provide a service cannot be excluded from the server candidates, and a server that is difficult to provide a service may be selected. It is difficult for a user who newly accesses such a server or another user who has been using the server before the user accesses the server to prevent deterioration of service quality due to lack of server resources.

Further, when the server having the function of not accepting a new server use request when the remaining resources of the server are not sufficient is the selected server, the server selected by the server load balancer accesses the insufficient server. The user cannot enjoy the service. The server selections shown above should be avoided.

Also, in particular, the above-mentioned article "Locating Nearb"
This section describes the problem of "Y Copies of Replicated Internet Servers." This method selects a server that can minimize the number of hops (the number of transit routers) between the server and the client, which has a great influence on the communication quality. However, if the nearest server does not have the remaining resources that can provide the service, the service quality cannot be maintained depending on the type of server or the server cannot accept the service use request and the user enjoys it. The quality of service that can be provided deteriorates.

On the other hand, Japanese Patent Laid-Open No. 11-41587 discloses a method of selecting a server having the smallest possible video transmission rate of the video server and the maximum number of simultaneously available channels.
According to this method, allocation can be performed in consideration of the remaining resources of the server, but since the required resource amount cannot be specified for each server selection processing request as in the above example, the remaining resources are required resources. It is possible that a server that does not exceed the above may be selected.

Further, as another problem, even when a server whose remaining resource exceeds the required resource is selected, since the distance from the user is not taken into consideration at all, a server having a large distance between the user and the server may be selected. May be selected. This increases the number of hops between the user and the server and causes quality deterioration on the network. As a result, it becomes difficult to maintain the quality of service for the user.

Next, a problem when a conventional server selection method is introduced for selecting a server for providing pay contents will be described. It is assumed that a business that provides paid content introduces a business model that charges and makes payment before enjoying the service.

In this case, it is not the client owned by the user who issues the server location resolution request to the server selection device, but an EC prepared by the business operator who provides the paid content.
( Electronic commerce ) server.
The EC server charges the content by controlling the interaction with the user. When the client accesses the URL of the server notified by the EC server after the end of billing, if the remaining resource amount of the server is not sufficient to provide the service, the service whose quality is deteriorated despite the billing is provided. There is a problem that you receive the service or you cannot receive the service at all.

Originally, a site that distributes paid contents needs to judge whether or not the server selected for the user can provide the service before charging. There is no server selection device provided. Further, it may take some time before the user actually accesses the selected server after the charging / settlement processing by the EC server. In order to prevent the former client from becoming inaccessible by another client accessing the server in the meantime, if the server has a temporary reservation function, the temporary reservation information is sent to the server selected when the server is selected. A function to notify is required.

However, the conventional server selection method does not have a means for notifying the information regarding the temporary reservation of the user by using the temporary reservation function of the server when the server is selected. It is also considered necessary to have a function of notifying the selected server of the user's authentication information in order to further strengthen security and provide services in a more limited form. This is also not possible because there is no means to notify the authentication information.

As a concrete example of a problem when the server selection device itself is decentralized, the server selection method of the server in which the number of remaining resources must be taken into consideration is described above.
Distribute Direc, a load balancer made by isco
The case where the server selection method based on the combination of tor and Local Calder is applied will be described.

As described above, in the conventional server selection method, the number of remaining resources of the server is compared with the required number of resources, and only the server having the remaining resources exceeding the required resources cannot be selected as the server selection candidate. Also in this example, the Local Director cannot select the server by comparing the remaining resources of the load-balanced server with the necessary resources, and further, the necessary resources are local to the Distribute Director.
There is no function to notify the Director. For example, Loca
l Although the service cannot be provided on all servers managed by the Director, the Distribute Direc
The tor may cause the Local Director to perform a server selection process, and as a result, a server that cannot provide the service may be selected.

These problems can be organized into the following four points. Problem 1: In the existing server selection method, there is a possibility that a server whose remaining resource amount is less than or equal to the resource amount required when providing a service may be selected. When a client accesses such a server, the processing of the entire server cannot be performed normally, so the quality of service of all users who use the server deteriorates, or the connection of the client is refused, and the service is You will either not be able to enjoy it.

Problem 2: There is a server selection method of selecting the server with the least server load, but since it does not consider the distance between the client and the server at all, it exists far from the client. The server may be selected. If the distance (the number of routers on the route, etc.) increases, delays and the like occur on the network. As a result, even if the server with the smallest server load is selected, the quality of service will deteriorate.

Problem 3: In the existing server selection method, even if the selected server has the provisional reservation function, this cannot be notified when the server is selected. If the selected server is not tentatively reserved, the server is accessed by another client between the time the client obtains the server selection result and the time the server is accessed, and the maximum number of simultaneous connections of the server at the current time is the maximum. When the number of accesses is reached, the access may be denied even if the client accesses the server.

Problem 4: In the existing server selection method, even if the selected server has an authentication function, this cannot be notified when the server is selected. In the conventional server selection method, if you select a server that requires user authentication as the selection target, there is only a method of distributing user authentication information to all servers that may be selected in advance, but in order to realize this Is premised on that all client information is registered in advance, and is not suitable for services that assume an unspecified number of users. Also,
This is inefficient because the same authentication data must be registered in all servers.

An object of the present invention is to solve the above-mentioned four problems, to provide a server selection technique which considers the amount of resources required for service provision and can select a server that is close to the user. To do. The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[0026]

The outline of the invention disclosed in the present application will be briefly described as follows.
The first invention is a means for solving the above-mentioned problems 1 and 2. A server selection device of the present invention includes a distance acquisition unit that acquires a user-server distance from user identification information and server identification information, and a remaining resource amount that acquires the remaining resource amount of the server from the server identification information. The acquisition unit, the user identification information, the server list, and the required server resource amount are input, and the remaining resource amount is acquired by the remaining resource amount acquisition unit for each server described in the server list, and the remaining resource is acquired. A server that preferentially selects a server having a short distance to the user obtained by the distance acquisition means from the servers having the required amount of server resources exceeding the amount and outputs the identification information of the server as the server selection result. And a selection means. By providing such a mechanism, it is guaranteed that the server selected by the server selection device of the present invention has a remaining resource amount that exceeds the required server resource amount and that the distance from the user is short. To be done.

A second invention is a one-distance acquisition method in the distance acquisition means of the first invention, and provides a concrete method for solving the above-mentioned problem 2. The distance acquisition means of the present invention includes a route information table storing identification information of relay devices existing on a route from the server selection device of the present invention to each server included in the server list, and the server selection device of the present invention. To the user for each of the servers included in the server list based on the relay device information processing unit that acquires the identification information of the relay device on the route from the user to the user, and the relay device information acquired by the relay device information processing unit. It is characterized by including a means for obtaining the distance of. By providing such a mechanism, both the route information from the server selection device to the user and the route information to the server can be acquired, so that the distance between the user and the server can be estimated and the server can be selected accordingly.

A third invention is a distance acquisition method in the distance acquisition means of the first invention, and provides a concrete method for solving the problem 2. A distance acquisition unit of the present invention includes a position acquisition unit that acquires position information of a user, a position table of a server that stores position information of each server, position information of a user acquired by using these, and a server position information. It is characterized in that it is provided with means for obtaining the distance between the server and the user using the position information. According to the present invention, even if the distance between the server selection device and the managed server or the user is large, the server-
Since the distance between the users can be accurately measured, it is guaranteed that the server selected by the server selection device is a server that can provide a service that is truly closest to the user.

A fourth aspect of the present invention is characterized in that the remaining resource acquisition means in the server selection device of the present invention includes a remaining resource management table indicating a remaining resource amount for each server that is a management target of the server selection device. This allows the server selection device of the present invention to perform resource allocation in consideration of the required resource amount even for a server that does not have a function of outputting the remaining resource amount in response to a request from an external device. The application area of the selection method expands.

In a fifth aspect of the invention, the maximum number of terminals that can be simultaneously connected to the server is defined as the maximum number of simultaneous connections, and the remaining resource amount is calculated from the maximum number of simultaneous connections of the server to the number of terminals connected to the server. It is characterized by the subtracted value. As a result, if the number of connected users has reached the maximum number of simultaneous connections defined by the server, it is determined that there are no remaining resources in the server, and this server is excluded from the selection candidates. It is possible to prevent the above client from connecting to the server.

According to a sixth aspect of the invention, the sum of the maximum transmission rates that the server can execute is defined as the maximum total rate, and the remaining resource amount is obtained by subtracting the sum of the data transfer rates being executed from the maximum total rate of the server. It is characterized by being a value. As a result, if the difference between the maximum total rate of the server and the data transfer rate that the server is currently executing is less than the data transfer rate that the server is about to start, it is determined that there are no remaining resources, and this server Is excluded from the selection candidates, it is possible to prevent the client from accessing the server that cannot secure the transfer rate required for providing the service.

According to a seventh aspect of the present invention, the total sum of rates of data transfer to and from outside the base that can be executed by a group of servers installed in the same base is defined as the maximum total base rate, and the remaining resource amount is
It is characterized in that it is a value obtained by subtracting the sum of the rates of data transfer with the outside of the site where the server group of the site where the server is installed is being executed from the maximum total rate of the site where the server is installed. As a result, it is determined that there is no remaining resource at the base where the difference between the maximum total base rate and the sum of the data transfer rates outside the base at the moment is lower than the data transfer rate necessary for providing the service. . Therefore, the server group arranged at the site where the transfer rate necessary for providing the service cannot be secured is excluded from the selection candidates.

In an eighth aspect of the present invention, the remaining resource amount of the server selected by the server selecting means out of the remaining resource amounts stored in the server resource management table in the fourth aspect is input to the server selecting means. It is characterized in that it has a remaining resource subtraction means for subtracting only required resources. As a result, even when the server managed by the server selecting means does not provide the information necessary for calculating the remaining resource amount, the remaining resource amount indicated by the server resource management table is the same as the actual remaining resource amount. Therefore, it is possible to realize the allocation considering the remaining resources.

In addition to the above-mentioned fourth invention, a ninth invention inputs the identification information of the service provided by the server and the server selected by the server selecting means, and provides the service based on this input. A service providing time acquiring unit that acquires a required time (service providing time) and outputs the service providing time, a service providing time output by the service providing time acquiring unit, and a server selected by the server selecting unit. After the identification information and the required resource amount are input, and after the service providing time has elapsed from the time when these are input, the remaining server resource amount related to the selected server indicated by the server resource management table is equal to the required server resource amount. Server resource management type that updates the server resource management table to increase It is characterized in that a chromatography unit. As a result, even when the server does not have the function of outputting the information on the remaining resources to the external device, it is possible to realize the server selection in consideration of the remaining resources of the server.

In a tenth aspect of the present invention, when the server to be managed by the server selection device has a provisional reservation function for temporarily securing server resources for the user's access, the server selection means is selected at the time of server selection. Input the identification information of the selected server, the identification information of the user, and the identification information of the service, which is the output of the above, and input the information for setting the temporary reservation for the server (hereinafter referred to as the temporary reservation setting information). Provision is made of provisional reservation information notifying means for generating based on these pieces of information, outputting the provisional reservation setting information, and outputting the identification information of the server selected after the provisional reservation setting is normally completed as the server selection result. It is characterized by As a result, the provisional reservation information of the user is notified only to the limited server, so that it is possible to reliably provide the service only to the limited user who accesses the limited server.

An eleventh aspect of the present invention is, when a server to be managed by the server selection device has a user authentication function for restricting user access, selects a server which is an output of the server selection means when selecting a server. Enter the server identification information, the user identification information, and the service identification information, and use the information (hereinafter referred to as the authentication setting information) for setting the user authentication for the user to the server based on these information. The present invention is characterized by including an authentication information notifying unit that generates the authentication setting information, outputs the authentication setting information, and outputs the identification information of the server selected after the authentication setting is normally completed as the server selection result. With this, since the user authentication information is notified only to the limited server, it is possible to reliably provide the service only to the limited user who accesses the limited server.

In a twelfth aspect of the invention, when the identification information of the user is input and whether or not the server can be selected by the server selection device is calculated based on the position information of the user, the selection is determined. In the case of continuing the server selection process by the server selection means and determining not to perform the selection, the identification information of the server selection device that is a candidate for the server selection for the user is output, and the server selection process is performed. It is characterized in that it is provided with means for determining whether or not the server can be interrupted. This makes it possible to decentralize the server selection program and execute the server selection operation. Hereinafter, embodiments of the present invention (Examples) will be described.
A detailed description will be given with reference to the drawings.

[0038]

1 is a block diagram showing a functional configuration of a server selection device according to an embodiment of the present invention, and FIG. 2 is a flowchart showing an operation procedure of the server selection device according to the present embodiment. As shown in FIG.
The server selection device 100 is composed of a server selection unit 1, a distance acquisition unit 2, and a remaining resource amount acquisition unit 3. In FIG. 1, U is user identification information, S is server identification information, SL is a server list (a list of server identification information), R is the required server resource amount, and SS is the server selected by the server selection means. Identification information, D
(US) is the distance information between the base U and the base S, R (S)
Is the remaining resource amount of the server S.

The operation of the server selection device 100 of this embodiment will be described with reference to FIG. First, the server selection device 100 of the present embodiment, when started, the user identification information U,
Server list SL and required server resource amount R
Is input to the server selection means 1 (101). The server selection means 1 obtains the remaining resource amount by using the remaining resource acquisition means 3 for all the servers described in the input server list SL (102). For a server whose remaining resource amount exceeds the required server resource amount R (103), the distance between the user and the server is obtained by the distance acquisition means 2 (104). The above operation is performed for all the servers described in the input server list. The above operation is performed for all the servers, and among the servers with the remaining resource amount exceeding the server resource amount, the server closer to the user is preferentially output as the selection result of the server selection device 100 (105). By performing such an operation, a server whose remaining resource amount is less than the required server resource amount is excluded from the server selection candidates. In addition, the server that is not excluded is selected as a result of the server selection by preferentially selecting a server having a short distance with the user, which is obtained by the distance acquisition unit 2, so that the selected server has the remaining resource amount. Is guaranteed to exceed the required server resource amount, and the distance from the user is short, so that the problem 1 and the problem 2 can be solved at the same time.

Next, the internal mounting structure of the server selection device according to the present embodiment and the environment surrounding the server selection device will be described. FIG. 3 is a block diagram showing the functional arrangement of a general-purpose computer having the server selection device according to the present embodiment of the present invention. As shown in FIG. 3, the general-purpose computer 50 having the server selection device according to the present embodiment of the present invention is connected to the communication line 52 having the router device 51. A managed server 53 of a server selection device is connected to the communication line 52, and a plurality of client devices 54 having client software 54A are connected to a predetermined router device 51. The internal configuration of the general-purpose computer 50 includes a server selection program 50A and a database 5 referred to by the server selection program 50A.
The server selection program 50A composed of 0B
Is a reception processing routine 50A1 and a main routine 50A
2 and a server selection routine 50A3.

UR with client software 54
The service specified by L is used as the server selection program 5
The protocol stack used when requesting 0A is shown in FIG.
Table 1 shows examples of messages used at that time. FIG. 4 and Table 1 show an example in which http and rtsp are used as the upper protocol in this protocol stack. Further, in FIG. 4, since the following description is independent of the data link layer and the physical layer, the layers below IP are omitted.

After selecting the server, the server selection device of the first embodiment notifies the client software of the server selection result using the protocol stack shown in FIG. Table 2 shows an example of upper-layer protocol messages at the time of result notification. The client software that received the message uses the character string "Lo described in the message.
The client device can access the target server by reflecting the server selection result by making a service use request to the URL in the line in which "cation:" is described. Transferring processing to the server is called redirection.

[0043]

[Table 1]

[0044]

[Table 2]

Further, the environment presupposed inside the server selection device will be explained in detail. As shown in FIG. 5, when the server selection program receives the service use request, it activates a child process that executes the main routine, causes it to perform all server selection processing, and the parent server process executes the client process again. Wait for the server selection request from.

A flowchart of the main routine is shown in FIG. When the main routine is activated, first, the IP address of the client device is acquired (201). Further, it is assumed that the information shown in Tables 3 and 4 is stored in the database, and the service U
The URL of the server corresponding to the RL, the URL of the service, and the resource amount necessary for providing the service are acquired (20
2). After obtaining these pieces of information, the server selection routine is activated (203). Main routine and server selection routine are common programming language and standard C / C ++
/ JAVA (registered trademark) or the like.
In such a case, the server selection routine is realized by the function shown in FIG. 7, and the server selection routine can be started by incorporating this function into the main routine or the like (20
4). The server selection routine has a function of changing the format of the server selection result to the format shown in Table 2 and notifying the client software of the server selection result.

[0047]

[Table 3]

[0048]

[Table 4]

An embodiment of the server selection program described above will be described below. (Embodiment 1) A server selection routine according to the above-mentioned first invention, wherein the above-mentioned second invention is used as the distance acquisition means,
An embodiment will be described in which the remaining resource amount recognized by the server selection routine is defined as in the above-mentioned fifth invention.

FIG. 8 shows an assumed network. Figure 8
Among them, SS 'is a server selection device in which the server selection program according to the present invention is installed. Ra, Rb, R
c, Rd, and Re are all general-purpose router devices. Figure 8
The numerical value described near the network connection part of the router device is used and the IP address of the router device is represented as Ra0 and Rb1. Sa, Sb, Sc, Sd,
Se is a server that is the selection target of SS '. Also,
U'is a user who makes a usage request for the server.

FIG. 9 is a functional block diagram of the server selection device realized in the first embodiment. In FIG. 9, 1 is a server selecting means,
2 is a distance acquisition unit, 3 is a remaining resource amount acquisition unit, U is a user IP address (user identification information), S is a server URL (server identification information), and SL is a server URL.
List (server identification information list), R'is 1 "connect"
(Required server resource amount), SS is URL of selected server (identification information of selected server), D (U.
S) is the distance information between the base U and the base S, and R (S) 'is the number of simultaneous connections possible at the present time.

As shown in FIG. 9, the server selection routine 100A of the first embodiment is composed of a server selection means 1, a distance acquisition means 2, and a remaining resource amount acquisition means 3. The server selection routine of the first embodiment is performed by the user's IP.
The address U and the URL list SL of the selected server are input, and the SS of the URL of the selected server is output. Also, the IP address U of the user and the URL list S of the server entered in the server selection routine 100A.
L and 1 “connection” R ′ are input to the server selection means 1. The server selection means 1 obtains the remaining resource amount acquisition means 3 for all the servers listed in the server URL list SL.
, The current number of possible simultaneous connections of each server R (S) '
To get.

The number of possible simultaneous connections R (S) ′ at present is 1
If “connection” R ′ or more, the distance acquisition unit 2 acquires the distance D (US) between the user and the server. The server selection means 1 selects, as an output of the server selection routine, a server having the shortest distance to the user from the servers in which the remaining resource amount exceeds the required server resource amount.

Details of each means will be described below. First, a specific example of the distance acquisition unit 2 will be described. As the first step, the route information between the server selection device and the user is acquired. Here, a program called traceroute, which is generally well known, is used as the route information acquisition means. The traceroute program is a program that, when the target IP address is input, outputs the IP addresses of the routers through which the packet reaches the IP address in order from the one closest to the program execution machine. By using the traceroute, the I of the transit router between the server selection device and the user
The P address can be obtained.

[0055] shows the traceroute execution results when the network configuration as described in FIG. 8 in FIG. 10. This is temporarily stored as the distance information of the user. As the second step, the server distance information is acquired for each server described in the server URL list.
In the first embodiment, as described in the above-mentioned second invention, it is premised that the distance acquisition means of the server selection device is provided with the route information table storing the identification information of the server selection device-relay device. Here, the relay device is a router,
The identification information is used as the router address. Table 5 shows an example in which the route information table is described in the network configuration shown in FIG . The route information table is defined as server distance information and is temporarily stored. As the third step, the distance information Lsu between the server and the user is obtained using the following equation 1 using the distance information of the user and the distance information of the server that are temporarily stored. In this case, the “distance” is the number of transit routers on the route.

[0056]

[Number 1] Lsu = A + B-2 × C +1 A = number C = server selection router in the path between the number B = server selection device and the server router in the path between the server selection device and user Table 6 shows the number of routers common to the route between the device and the user and the route between the server selection device and the server . This is defined as distance information between the server and the user and is temporarily stored.

[0057]

[Table 5]

[0058]

[Table 6]

The remaining resource amount acquisition means 3 is the UR of the server.
Let L be the input and the number of possible simultaneous connections of the server at the present time be the output. The current number of possible simultaneous connections of the server shall be acquired by inquiring to the server.
The remaining resource amount acquisition means 3 is provided by a library file and an API for inquiring the server.

The server selection means 1 uses the temporarily stored distance information between the server and the user, and uses the remaining resource amount acquisition means 3 for the server in which the distance information between the server and the user is described. Get a resource. Further, in the first embodiment, the remaining resource amount is a value obtained by subtracting the number of terminals connected to the server from the maximum number of simultaneous connections defined in the fifth aspect of the invention, and is required for one server processing request. Since the resource amount is "1 connection", the server selection means 1 may select the nearest server from the ones in Table 7 where the remaining resource amount is not 0. Therefore, when the remaining resource amount is as shown in Table 7, the URL of the server selected by the server selecting means 1 is the URL of Sa. This URL is the output URL of the server selection program
Becomes

[0061]

[Table 7]

As shown above, the server selection program having such a mechanism can select the server closest to the user and capable of accommodating the user from the plurality of server lists.

(Second Embodiment) A second embodiment according to the present invention in which the resource to be handled is a server band as shown in the sixth invention will be described below by modifying the first embodiment. The presumed network configuration is the same as that shown in FIG . Also,
FIG. 11 shows a functional block diagram of a server selection routine realized in the second embodiment. In FIG. 11, 100B is a server selection routine of the second embodiment, 1 is server selection means, 2 is distance acquisition means, 3 is remaining resource amount acquisition means, U is the IP address of the user, S is the URL of the server, and SL is SL. Server URL list, R '' is the network bandwidth required for the server,
SS is the URL of the selected server, D (U.S) is the distance information between the base U and the base S, and R (S) ″ is the maximum possible transmission rate at the present time. As shown in FIG. 11, the part different from the first embodiment is a server selection routine 100B.
Is that a band R ″ necessary for providing a service is added as an input of the above, and the remaining resource is the maximum possible transmission rate R (S) ″ at the present time of the server.

The bandwidth R ″ necessary for providing the service is “bi
It is input from the main routine to the server selection routine in the unit of "t / sec". The server selection means uses the remaining resource amount acquisition means 3 for each server described in the input URL list SL of the server, and the maximum transfer rate possible. R (S) ″ is obtained. Table 8 shows an example of the remaining resource amount of the server. The maximum transfer rate R (S) ″ required for each server and the bandwidth R ″ required for service provision are shown. A comparison is performed, and a server in which R (S) ″ is lower than R ″ is excluded from the server selection candidates.

[0065]

[Table 8]

The other processing is the same as that of the first embodiment, and therefore its explanation is omitted. As described above, the server selection program having such a mechanism can select the server closest to the user and capable of accommodating the user from the plurality of server lists.

(Embodiment 3) A server device according to the above-mentioned first invention, wherein the distance acquisition means uses the above-mentioned third invention, and the remaining resource amount recognized by the server selection device is determined by the above-mentioned seventh invention. A third embodiment in which the above definition is made will be described. FIG. 12 shows an assumed network. FIG. 12 shows a state in which the bases indicated by Aa, Ab, Ac, Ad, Ae, As, Au are connected via the router device.

FIG. 13 shows a functional block diagram of the server selection routine realized in the third embodiment. The server selection routine 100C of the third embodiment inputs the site name to which the user belongs, the URL list of the selection target server, and the available bandwidth required for the line extending from each site to the backbone, and outputs the URL of the selected server. . The service selection routine 100C includes a server selection unit 1, a distance acquisition unit 2, and a remaining resource amount acquisition unit 3.

First, the distance acquisition means 1 will be described.
In the first step, the IP address of the user is used, and a numerical value is calculated by relativizing the position of the user when considering the topology in the network. Here, the IP address is x. y.
z. When expressed as w (0 <x, y, z, w <256), it is assumed that a numerical value obtained by relativizing the position of the user is obtained by the following formula 2 and temporarily stored.

[0070]

[Formula 2] (x & m1) x 256 ³ + (y & m2) x 256 ² +
(z & m3) × 256 + (w & m4) However, the net mask is m1. m2. m3. m4 (0 ≦ m1, m2, m3, m
Expressed as 4 <256). It is defined that a & b means a logical sum for each bit. That is, a = 0x15, b
If = 0xbb (0x indicates a hexadecimal number), then a & b = 0x11. In the second step, the IP address is obtained from the URL of the server, the numerical value obtained by relativizing the position of the server is obtained based on the above equation, and is temporarily stored. In the third step, the distance between the user and the server is obtained from the temporarily stored numerical value of the relative position of the user and the numerical value of the relative position of the server, and this is temporarily stored. . Table 9 shows the relativized numerical values of the server when the IP address and the subnet mask of the server shown in Table 9 are assumed. It is also assumed that the user's IP address is 11.14.5.12 and the subnet mask is 255.255.255.0. In this case, the user's relativization numerical value is 185794560.

[0071]

[Table 9]

Next, the remaining resource amount acquisition means 3 will be described. Table 10 shows a table held by the remaining resource amount acquisition means 3. The third embodiment is an example in which the resource is a band of a network line from the base. The table shown in Table 10 is characterized in that it is not held for each server but is managed for each base.

[0073]

[Table 10]

Next, the server selecting means 1 will be described. The server selection means 1 has a table as shown in Table 11 that associates the servers with the bases. After calculating the distance between the server and the user by the following formula 3 based on the table such as Table 11 and the information of the relative value of the server / user, and after checking whether the server has the remaining resources Among the servers that can provide services, the one with the shortest distance to the user is output as the server selection program.

[0075]

## EQU00003 ## | (numerical value relative to user)-(numerical value relative to server) | Table 12 shows the relative numerical value of the user and the relative numerical value of the server and the available bandwidth shown in Table 10. The table described above is shown. From Table 12, this server selection program has the shortest distance and the free bandwidth is the required bandwidth (here, 6 Mbit / se.
It can be seen that those satisfying (denoted as c) are servers of Sb0 and Sb1. The server selecting means 1 may select a plurality of servers at the same base as in the above example, but in the third embodiment, a random number corresponding to each server is obtained and a server having a large random value is selected. Shall be decided. That is, in the above example, the random number is obtained twice,
The first result is Sb0 and the second result is Sb0.
As a result of 1, the one having a large random value is determined as the selection server. The random number has fixed minimum and maximum values and is uniformly distributed within the range.

[0076]

[Table 11]

[0077]

[Table 12]

With the server selection program having the above-described mechanism, it is possible to select the server closest to the user and capable of accommodating the user from the plurality of server lists.

(Embodiment 4) In order to deal with the case where the server to be managed by the server load balancing program does not have a means for notifying the external device of the remaining resource amount at the present time, the above-mentioned fourth invention, A fourth embodiment in which the eighth invention and the ninth invention are applied will be described. The fourth embodiment is assumed to be an extension of the first embodiment. That is, the input of the server selection routine is the IP address of the user and the URL list of the server, and the output is the URL of the selected server. Also, the amount of server resources that needs to be handled by the server selection routine is the number of connections to the server, and is one connection.

FIG. 14 shows the functional arrangement of the server selection routine of the fourth embodiment. In FIG. 14, 100D is a server selection routine of the fourth embodiment, 1 is server selection means, 2 is distance acquisition means, 3 is remaining resource amount acquisition means, 4 is server resource management timer means, U is the IP address of the user, S is the URL of the server, SL is the URL list of the server, R'is 1 "connect", SS is the URL of the selected server, D
(US) is the distance information between the base U and the base S, R
(S) 'is the number of simultaneous connections possible at the moment, TS is a timer start notification (notifying the URL of the selected server), RSS is a remaining resource subtraction notification (notifying the URL of the selected server), and RAS is the remaining resource. It is a notification of addition (notifying the URL of the selected server). As shown in FIG. 14, the server selection routine 100D according to the fourth embodiment includes a server selection unit 1, a distance acquisition unit 2, a remaining resource amount acquisition unit 3, and a server resource management timer unit 4. Here, it is assumed that the distance acquisition unit 2 is the same as the distance acquisition unit described in the first embodiment.

Remaining resource acquisition means 3 in the fourth embodiment
Will be described. First, the remaining resource acquisition means 3
Contains a server resource management table. The server resource management table is described in relation to the server name and the number of possible connections. Table 13 shows an example of the server resource management table. The table itself is realized using a database or the like.

[0082]

[Table 13]

Further, the remaining resource acquisition means 3 provides other function blocks with a function for acquiring and changing the contents of the server resource management table. C
Table 14 shows an example of functions assuming that programming is performed in the / C ++ / JAVA programming language.

The server selecting means 1 in the fourth embodiment will be described. FIG. 15 shows a processing flow of the server selection means. As a premise, it is assumed that the server list input to the server selection program and the user-server distance information obtained by the distance acquisition unit 2 are temporarily stored. As a first step, the temporarily stored server list is referred to, and the number of possible connections of each server described in the server list is obtained from the remaining resource amount acquisition means 3 (3
01).

[0085]

[Table 14]

The acquisition method is getReso described in Table 14.
urce function is used. In the second step, the server closest to the user is selected from the servers having the number of possible connections of 1 or more based on the temporarily stored user-server distance information (302). The part corresponding to the second stage is the same as the method described in the first embodiment. As a third step, a remaining resource subtraction instruction for the selected server is issued to the remaining resource amount acquisition means 3 (303). The subtraction instruction is the deleteResource described in Table 14.
Use a function. As a fourth step, the server resource management timer means starts a timer process for the selected server (304). The server selected as the fifth stage is output (305), and the processing of the server selecting means 1 is terminated.

The server resource management timer means 4 in the fourth embodiment will be described. FIG. 16 shows a processing flow of the server resource management timer means 4. As a first step, the server resource management timer 4 is started in the fourth step of the server selecting means 1 (401). The URL of the selected server is notified at startup. As the second step, the time required to provide the service is acquired (40
2). As a premise, the server resource management timer means has an estimated service provision time table showing the relationship between the URI, the protocol, and the estimated service provision time estimated to be required for service provision. Table 15 shows an example of this estimated service provision time table. This estimated service provision time table is implemented using a database or the like.

[0088]

[Table 15]

Further, in order to express the variation in the estimated service provision time due to the processing capacity of the server and the protocol used, the server resource management timer means 4 describes the relationship between the server name, the protocol and the processing capacity coefficient. It has a processing capacity coefficient table. By multiplying the estimated service provision time described in the estimated service provision time table by the processing capacity coefficient, the estimated service provision time in the selected server and the protocol used can be acquired. Table 16 shows a server processing capacity coefficient table.

[0090]

[Table 16]

The table shown above is prepared, the URL of the selected server is decomposed into a protocol, a server name, and a URI, the estimated service provision time is acquired from the decomposed protocol and URI, and processing is performed from the server name and the protocol. The estimated service provision time for the URL can be obtained by obtaining the capacity coefficient and multiplying the estimated service provision time by the processing capacity coefficient. As an example, UR
L is rtsp: //osaka.ntt.co.jp/ne
Assuming ws / 0705 / special.mpg, the protocol is rtsp and the server name is saka.ntt.c.
o.jp, URI is / news / 0705 / speci
Since it is al.mpg, the estimated service provision time for URI is 1 hour 5 minutes 30 seconds from Table 15,
The processing capacity coefficient is 1.5 from Table 16. Therefore,
The estimated service provision time for this URL is 1 hour 38 minutes 15 seconds.

As the third step, the time to wait until the next process is performed after the server resource management timer means 4 shifts to the waiting state is determined (403). Usually the UR
Although it is the same value as the estimated service provision time for L, it is also possible to add an arbitrary margin value. Example 4
In this case, a 10% margin is added. That is, when the estimated service provision time is 1 hour 38 minutes 15 seconds, the waiting time is calculated to be 1 hour 48 minutes 5 seconds (less than 1 second is advanced). As the fourth step, the execution of the process of the next fifth step is suspended for the waiting time.
(404).

In the fifth stage, after the waiting time,
The remaining resource amount acquisition means 3 is instructed to add the remaining resources of the selected server (405). The addition instruction is executed by the addResource function shown in Table 14. When the process of the fifth step is finished, the process of the server resource management timer is finished. Note that the server resource management timer process can execute a plurality of timer processes in parallel. In the fourth embodiment, it is possible to realize server selection in consideration of the remaining resources of the server without inquiring the server about the status of the remaining resources.

(Fifth Embodiment) The tenth invention and the eleventh invention described above are applied in order to deal with the case where the server to be managed by the server load balancing program has the provisional reservation function and the user authentication function. Example 5 in the case will be described.

First, the definition of "temporary reservation" of the server is made. The provisional reservation of the server means that when the server provides a service to a specific user, the server resource necessary for providing the service is secured in advance, and the server resource is used only by the specific user. Is an operation that guarantees.

Next, the "user authentication" of the server will be defined. "User authentication" is to verify whether the user who accessed the server is the user who is actually given the right to use the server. If the user is given the right, the service is provided and the right is given. If the user is not authorized, the function rejects the user's request.

Since the server has the function of executing the temporary reservation and the user authentication defined above, the service can be surely provided only to a limited number of users.
A fifth embodiment using the tenth invention and the eleventh invention in order to realize a server selection program for selecting a server having a provisional reservation / user authentication function will be described below. It is assumed that the fifth embodiment is positioned as the function expansion of the first embodiment.

First, before the description of the server selection program according to the fifth embodiment, the premise of the server that is the selection target of the server selection program will be described. It is assumed that the server to be selected is a server whose maximum number of simultaneous connections is determined by the number of software licenses. In addition, the provisional reservation function that reserves the resource for the user indicated by the IP address by specifying the user's IP address and the URI and protocol of the service to be used, and the user's IP address and the URI and protocol of the service to be used. It is assumed to have a user authentication function for authenticating the user who is designated and indicated by the IP address. A library is provided to control the server from a remote device, and the API
It shall be possible for the server selection routine to include a server control program through the (Aplication Replication Interface).

FIG. 17 shows the relationship between the library and the server selection routine. In FIG. 17, 100D 'is a server selection program, 100D is a server selection routine, 10 is a server control library, 20 is a selection target server group, and 20A.
Is a server to be selected. Table 17 shows the functions (temporary reservation function, authentication information function) provided by the library.

[0100]

[Table 17]

FIG. 18 shows the functional arrangement of the server selection routine of the fifth embodiment. In FIG. 18, 100E is a server selection routine of the fifth embodiment, 1 is server selection means, 2 is distance acquisition means, 3 is remaining resource amount acquisition means, 5 is temporary reservation information notification means, 6 is authentication information notification means, U is the IP address of the user, S is the URL of the server, SL is the URL list of the server, R'is 1 "connect", SS is the UR of the selected server.
L and D (US) are distance information between the base U and the base S,
R (S) 'is the current number of simultaneous connections possible, IRA is the information notification result from the provisional reservation information notifying means, and IRB is the information notification result from the authentication information notifying means.

The server selection routine 100E of the fifth embodiment.
As shown in FIG. 18, includes a server selecting unit 1, a distance acquiring unit 2, a remaining resource amount acquiring unit 3, a temporary reservation information notifying unit 5, and an authentication information notifying unit 6. The distance acquisition unit 2 and the remaining resource amount acquisition unit 3 are the same as those in the first embodiment.
Same as. In the fifth embodiment, the resources handled by the server selection routine are the same as those in the first embodiment. That is, it is assumed that the remaining resource of the server is the maximum possible number of connections at the present time, and the amount of server resources required for one service provision is one.

Each functional block will be described below.
First, the server selection means 1 will be described. FIG. 19
A flowchart of the processing procedure of the server selecting means 1 of the fifth embodiment is shown in FIG. Since the first step and the second step are the same as those described in the fourth embodiment, the description thereof will be omitted (501, 502). In the third stage, the URL of the server and the user's IP selected for the provisional reservation information notifying means 5
The address is notified (503). Provisional reservation information notification means 5
It is checked whether the processing result of No. is successful in the temporary reservation setting (504), and when the processing result of the temporary reservation information notifying unit 5 indicates that the temporary reservation setting has failed (NO), this server selection The processing in the means 1 is stopped (505).
When the processing result indicates that the temporary reservation setting is successful (Y
es), and shifts to the next fourth stage. In the fourth stage, the URL of the selected server and the IP address of the user are notified to the authentication information notifying means 6 (506). Next, it is checked whether the processing result of the authentication information notifying means 6 is successful in setting the authentication information (507), and if it indicates that the setting has failed (NO), the provisional reservation setting is canceled (508),
The processing in the server selecting means 1 is stopped (50
9). When the processing result indicates that the authentication information setting has succeeded (Yes), the process proceeds to the next fifth step. In the fifth stage, the URL of the selected server is output (510), and the processing of the server selecting means 1 is ended.

The provisional reservation information notifying means 5 will be described. FIG. 20 shows a processing flow of the provisional reservation information notifying means 5. URL sent from the server is rtsp: // tok
yo.ntt.co.jp/news.mp2. First
At the stage, the type of the server used for providing the service is specified from the URL notified from the server selection means 1 (6
01). It is assumed that the temporary reservation information notifying means 5 is provided with a correspondence table of server types, protocols, and file names shown in Table 18. It should be noted that * in the table indicates that the character string consists of one or more characters.

[0105]

[Table 18]

In the case of the example of the URL, the server type is specified as mpeg2_server according to Table 18. In the second stage, the provisional reservation notification program corresponding to the specified server is activated (602). Provisional reservation information notification means 5
Describes a correspondence table between server types and provisional reservation notification program names as shown in Table 19. In the third stage, the execution result of the temporary reservation setting is output, the URL of the selected server is output (603), and the processing of the temporary reservation information notifying means 5 is ended.

[0107]

[Table 19]

Further, as an example, the program mp2Pre
The processing sequence of Reserve is shown in FIG. First, the server name and URI are acquired from the URL (70
1). When mpeg2_server provides the function as shown in Table 17, the information necessary for the temporary reservation setting is the server name, URI, user address, and expiretime. Here, "expiretime" is defined as a time during which the tentative reservation is valid, and it is assumed that the tentative reservation is canceled if there is no access after this time after notification of the tentative reservation information.

The expirtime is the same as in the fifth embodiment.
In, fix at 1 minute. Obtain the server URI and server name from the server URL. Using the URI, server name, and user address as arguments, setPre in Table 17
The Reserve function is executed (702). The function setPre described in Table 17 together with this and the user address.
Notify as an argument of Reserve. The result of this execution is notified (703), and the execution of mp2PreReserve ends. In addition, all the programs shown in Table 19 are “(OS prompt) program name URL
It is assumed that it is activated in the form of "user address". The output values of the execution result are as shown in Table 20.

[0110]

[Table 20]

In the third stage, the output value of the execution result is directly used as the output of the temporary reservation notifying means 5, and the processing of the temporary reservation notifying means 5 is terminated. Next, the authentication information notifying means 6 will be described. The processing flow of the authentication information notifying means 6 is shown in FIG. Rt the URL notified by the server
sp: //tokyo.ntt.co.jp/news.mp
Set to 2. In addition, the IP address of the user is 10.2.3.
Set it to 0.100. In the first stage, the type of the server used for providing the service is specified from the URL notified from the server selecting means 1 (801). It is assumed that the authentication information notifying means 6 is provided with a correspondence table of server types, protocols, and file names shown in Table 21. The table shows that any character string is included.

[0112]

[Table 21]

In the case of the example of the URL, the server type is specified as mpeg2_server according to Table 21. In the second stage, the authentication information notification program corresponding to the specified server is activated (802). In the third stage, the execution result of the authentication information is output, the URL of the selected server is output (803), and the execution of mpeg2_server is terminated. The authentication information notifying means 6 describes a correspondence table between server types and authentication information notifying program names as shown in Table 22.

[0114]

[Table 22]

Also, as an example, the program mp2Aut
FIG. 23 shows a processing sequence of the entication. First, get the server name and URI from the URL (9
01). When mpeg2_server provides the command as shown in Table 17, the information necessary for provisional reservation setting is the server name, URI, user address, user name, and password. The URI of the server and the server name are acquired from the URL of the server and used as the argument of the function described in Table 17. Further, the user name and password are acquired from the user address (902). In the fifth embodiment, a user management database (DB) is provided in addition to the server selection device 100, and the user management database (DB) describes the relationship between the user address, the user name, and the password. And Together, these are used as arguments of the command setAuthentication described in Table 17, execute setAuthentication (903), and notify the authentication information. This setAut
Hentification execution result is output (90
4), execution of mp2Authentication ends. In addition, all the programs shown in Table 22 are
It shall be started in the format of "(OS prompt) program name URL user address". The output value of the execution result is as shown in Table 23.

[0116]

[Table 23]

In the third step, the output value of the execution result is directly used as the output of the authentication information notifying means 6, and the processing of the authentication information notifying means is ended. In the fifth embodiment, it is possible to configure a server selection routine corresponding to the case where the server has the provisional reservation function / user authentication function. This can be expected to bring the following merits. -The URL passed to the user as a result of server selection is guaranteed that the user can receive the service.
It is RL. Even if there is some time lag before the user accesses the server after selecting the server, the resources allocated to the user are not occupied by other users, so that the user can reduce the time after selecting the server. The user can certainly enjoy the service even when he / she accesses the server without permission.

(Embodiment 6) An embodiment using the above-mentioned twelfth invention when the server selection programs are distributed in order to cope with an increase in the load on the server selection device itself in which the server selection program is installed. 6 is shown.
FIG. 24 shows a presumed network configuration. Figure 24
Among them, SS ₁ and SS ₂ are server selection devices, Ra to Re are router devices, Sa to Se are selection target servers, and U is a user. Further, there are two regions (gray regions) surrounded by dotted lines, one is the management range of SS ₁ and the other is the management range of SS ₂ . As another premise of the sixth embodiment, the sixth embodiment is realized by expanding the first embodiment.

FIG. 25 shows the functional arrangement of the server selection routine of the sixth embodiment. Server selection routine 1 of the sixth embodiment
00F is the server selection routine 100A of the first embodiment.
And server selectability determination means 7. Example 6
FIG. 26 shows the processing flow of this server selection routine.
In the server selection routine 100F of the sixth embodiment, as shown in FIG. 26, the user selects the server selection routine 100F.
Determine whether it is appropriate to select the server in A (1
001), as a result, when it is determined that it is more appropriate to leave it to another server selection device (NO), the URL of the server selection device that seems to be more suitable for server selection for the user is obtained (1002), The URL of the server selection device is output (1003). In addition, when it is determined that it is suitable for the user to select the server in the server selection routine 100F, the determination result is Yes (Yes).
s), the server selection routine 100A of the first embodiment is activated, and the URL indicating the server selection result is obtained (100
4) Output the URL of the server selection result (100
5). As a result, the execution of the server selection program ends.

Now, the server selectability judgment means 7 in the sixth embodiment will be described. First, the server selectability determination means 7 is provided with a table that holds route information up to an area in which processing is left to another server selection device in advance. Table 24 shows an example in which a table holding the route information up to the area to which the processing is given is described in the network configuration shown in FIG.

[0121]

[Table 24]

The server selectability judging means 7 executes traceroute when receiving the IP address of the user. If the execution result of the traceroute includes any one of the address lists of the routed routers described in the route information holding table described in Table 24, the URL of the redirect destination server selection device described in the table is output to select the server. The permission / prohibition determination process is terminated. Here, a case where the user U requests the server selection from SS ₁ will be described as an example. First, the trace from SS ₁ to user U
Run route. Table 25 shows the execution results of the traceroute.

[0123]

[Table 25]

Since the result of Table 25 includes the route information in the first line of Table 24 (1), the server selection process is not performed in SS ₁ , and the redirect destination of the redirect line described in the first line of Table 24 (1) is not executed. The value of SS _2, which is the server URL, is output. Next, a case where the user requests SS ₂ to select a server will be described. First, the trac from SS ₂ to U
Execute the route. Table 26 shows the result of executing the traceroute.

[0125]

[Table 26]

Since this result does not include any of the route information in Table 24 (2), it can be determined that the user is a user who should select a server with this server selection device. Therefore, the server selection program described in the first embodiment is activated to output the selected server. According to the sixth embodiment, it is possible to decentralize the server selection program and execute the server selection operation using the present invention.

(Embodiment 7) In this embodiment 7, assuming that the server selection device of the present invention is used to provide a paid service, an EC ( electronic
It will be described cooperation example of a click-commerce) server.
As a premise, a service for streaming delivery of moving image content is assumed. There are a plurality of moving image distribution servers, and the maximum number of simultaneous connections is determined by the number of server software licenses. When the user accesses the free home page corresponding to the content catalog and selects the content that he / she wants to watch for a fee, the information necessary for charging is interactively exchanged between the EC server and the user terminal, and after the charging process is completed, the user The EC server notifies the URL for browsing the pay content. In addition, the server selection device includes
The server selection program of Also,
It is assumed that the name of the user and the personal identification number of the user are registered in the EC server.

The overall sequence of service provision is shown in FIG.
7 shows. The system that provides this service is a client prepared by a user who intends to receive paid contents, and an E provided by an information provider who provides paid contents.
EC server database that stores C server and customer information, server selection device that selects a server that can provide a service from a plurality of moving image servers that is close to the client, and a movie that is a selection target of the server selection device It consists of an image distribution server. According to the sequence shown in FIG. 27, the server resource of the user is tentatively reserved immediately before the contract is concluded (agreement to the billing) between the user and the information provider, so that the resource may not be available. It is possible to avoid troubles such as not being charged, or not being able to enjoy the service when accessing the URL notified after the charging, and it is possible to provide a more reliable pay service. Further, since the user authentication information is notified only to the server selected by the server selection device, it is not necessary to previously place a copy having the same contents as the EC server database in all the servers, and the server cost can be reduced.

As described above in the present embodiment, the server selection device using the present invention can be applied to a server that provides a pay service, and can easily realize cooperation with an EC server that performs billing settlement. The processing program of each of the above embodiments is recorded in a computer-readable recording medium,
You may make it use again. Here, the "computer-readable recording medium" means a floppy (registered trademark) disk, a magneto-optical disk, a ROM, a CD-ROM.
A portable medium such as a hard disk or a recording device such as a hard disk built in a computer system. The invention made by the present inventor has been specifically described above based on the embodiment (example), but the present invention is not limited to the embodiment (example).
Of course, the present invention is not limited to the above, and various changes can be made without departing from the scope of the invention.

[0130]

The effects obtained by the invention disclosed in the present application will be briefly described as follows.
According to the present invention, when a plurality of server devices provide the same service on the communication network and the user intends to use the service provided by the server group, the plurality of server devices are to be accessed by the user. You can surely secure the resources required for service provision from the server device,
In addition, it is possible to select a server that is close to the user.

[Brief description of drawings]

FIG. 1 is a functional configuration diagram of a server selection device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an operation procedure of the server selection device of this embodiment.

FIG. 3 is a functional configuration diagram of a general-purpose computer having a server selection device according to an embodiment of the present invention.

FIG. 4 is a diagram showing a protocol stack used between a client device and a server selection device according to the present embodiment.

FIG. 5 is a diagram for explaining an environment premised on the inside of the server selection device of the present embodiment.

FIG. 6 is a flowchart showing a processing procedure of a server selection program of this embodiment.

FIG. 7 is a diagram showing an implementation example of a server selection routine of the present embodiment.

FIG. 8 is a diagram showing a network assumed in the first embodiment according to the present invention.

FIG. 9 is a functional configuration diagram of a server selection routine in the first embodiment.

FIG. 10 is a traceroute in the first embodiment.
It is a figure which shows the example of a result of execution.

FIG. 11 is a functional configuration diagram of a server selection routine in a second embodiment according to the present invention.

FIG. 12 is a diagram showing a network assumed in a third embodiment according to the present invention.

FIG. 13 is a functional configuration diagram of a server selection routine in the third embodiment.

FIG. 14 is a functional configuration diagram of a server selection routine according to a fourth embodiment of the present invention.

FIG. 15 is a flowchart illustrating a processing procedure of a server selection unit according to the fourth embodiment.

FIG. 16 is a flowchart showing a processing procedure of a server resource management timer means of the fourth embodiment.

FIG. 17 is a diagram showing a relationship between a server selection routine and a server control library according to the fifth embodiment of the present invention.

FIG. 18 is a functional configuration diagram of a server selection routine in the fifth embodiment.

FIG. 19 is a flowchart showing a processing procedure of a server selection means of the fifth embodiment.

FIG. 20 is a flowchart illustrating a processing procedure of a provisional reservation information notification unit according to the fifth embodiment.

FIG. 21 is a flow chart showing a processing procedure of an mp2PreReserve program activated by the provisional reservation information notifying means of the fifth embodiment.

FIG. 22 is a flowchart showing the processing procedure of the authentication information notifying means of the fifth embodiment.

FIG. 23 is a flowchart showing a processing procedure of an mp2Authentication program activated by the authentication information notifying means of the fifth embodiment.

FIG. 24 is a diagram showing a network assumed in a sixth embodiment according to the present invention.

FIG. 25 is a functional configuration diagram of a server selection routine of the sixth embodiment.

FIG. 26 is a flowchart showing a processing procedure of a server selection routine of the sixth embodiment.

FIG. 27 is a diagram illustrating a sequence among a client, an EC server, a server selection device, and a moving image distribution server, which is assumed in a seventh embodiment according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Server selection means 2 ... Distance acquisition means 3 ... Remaining resource amount acquisition means 4 ... Server resource management timer means 5 ... Temporary reservation information notification means 6 ... Authentication information notification means 7 ... Server selectability judgment means 10 ... Server control library 20 Selection target server group 20A Selection target server 50 General-purpose computer 50A Server selection program 50B Reference database 51 Router 52 Communication line (network) 53 Server managed server management server 54 Client device 54A Client software 100 ... Server selection devices 100A to 100F ... Server selection routine U'of the present embodiment ... User U requesting to use the server U ... User identification information S ... Server identification information SL ... Server list (server identification information List) R ... Amount of required server resources SS '... Server device SS in which the server selection program is installed ... URL (identification information) of the selected server D (US) ... Distance information R (S) between base U and base S ... Remaining resource amount SSY of server S ... Server identification information SSN when selectable ... Server identification information SY when not selectable ... Message indicating selectable

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-11-110324 (JP, A) JP-A-7-152698 (JP, A) JP-A-7-160648 (JP, A) JP-A-11- 55645 (JP, A) JP 10-78938 (JP, A) JP 2000-196677 (JP, A) International publication 00/28713 (WO, A1) International publication 99/05094 (WO, A1) (58) Fields surveyed (Int.Cl. ⁷ , DB name) G06F 13/00 G06F 15/16 H04L 12/00

Claims (23)

(57) [Claims] 1. A server selection device for selecting a server to be accessed by a user from among a plurality of distributed servers, wherein the user identification information and a server list which is a list of the servers are input, Distance acquisition means for acquiring the distance to the user for a plurality of servers,
The remaining resource amount acquisition means for acquiring the remaining resource amount of the server for the predetermined server, the identification information of the user, the server list, and the required server resource amount are input and included in the server list. And the remaining server resource amount acquired by the remaining resource amount acquisition means exceeds the required server resource amount, and the distance to the user indicated by the input obtained by the distance acquisition means Server selecting means for preferentially selecting a server with a short server and outputting the identification information of the server as the server selection result ;
If you have a user authentication function to limit access
When the server is selected, it is the output of the server selecting means.
Selected server identification information, user identification information, and
Enter the identification information of the service,
Information for setting user authentication for users (below
Lower authentication setting information) based on these information
Output the certificate setting information and select it after the authentication setting is completed normally.
Outputs the identification information of the selected server as the server selection result
A server selection device, comprising: 2. The distance acquisition means acquires, for each of the plurality of servers, a route information table that stores identification information of a relay device on the route from the server selection device to the server, and the identification information of the relay device. 7. A relay device information acquisition means, and a distance calculation means for calculating a distance to the user for each of the plurality of servers based on the relay device identification information acquired by the relay device information acquisition means. 1. The server selection device according to 1. 3. The distance acquisition unit, a position acquisition unit that acquires position information of the user from identification information of the user, a server position table that stores position information of the server for each of the plurality of servers, and Characterized in that it includes means for obtaining the distance to the user for each of the servers based on the identification information of the user acquired by the position acquisition means and the position information of each of the servers stored in the server position table. The server selection device according to claim 1. 4. The remaining resource amount acquisition means includes a server resource management table that stores remaining resource information indicating a remaining resource amount for each of the plurality of servers. The server selection device according to item 1. 5. The number of terminals that can be simultaneously connected to the server is defined as the maximum number of simultaneous connections, and the remaining resource amount is the maximum number of simultaneous connections of the server to the number of terminals connected to the server. 5. The server selection device according to claim 1, wherein the server selection device has a value obtained by subtracting. 6. A total sum of data transfer rates that the server can execute is defined as a maximum total rate, and the remaining resource amount is the maximum total rate of the server to the rate of data transfer being executed by the server. any one of the claims 1 to 5, characterized in that a value obtained by subtracting the sum of
A server selection device according to item. 7. The sum total of the rates of data transfer to and from outside the sites that can be executed by a group of servers installed at the same site is defined as the site maximum total rate, and the remaining resource amount is the value of the site where the server is installed. 5. A value obtained by subtracting the sum of rates of data transfer to / from outside the bases in which the server group of the base where the server is installed is subtracted from the maximum total base rate. The server selection device described. 8. A remaining resource subtraction for subtracting, from the remaining resource amount stored in the server resource management table, the remaining resource amount of the server selected by the server selecting unit by the necessary resource input to the server selecting unit. The server selection device according to claim 4, further comprising means. 9. The identification information of the service provided by the server and the server selected by the server selection means are input, and the time required for the service provision (hereinafter referred to as the service provision time) is acquired based on this input. , A service providing time acquiring unit that outputs the service providing time, the service providing time output by the service providing time acquiring unit, the identification information of the server selected by the server selecting unit, and the required resource amount Then
After the service providing time has elapsed from the time when these were input, the server resource management table is updated so that the remaining server resource amount related to the selected server indicated by the server resource management table increases by the required server resource amount. The server selection device according to claim 4, further comprising server resource management timer means. 10. When the server to be managed by the server selection device has a provisional reservation function for temporarily securing server resources for the user's access, the output of the server selection means is made when the server is selected. The identification information of the selected server, the identification information of the user, and the identification information of the service are input, and the information for setting the temporary reservation for the server (hereinafter referred to as temporary reservation setting information) is provided as these information. A provisional reservation information notifying unit having a function of generating the original, outputting the provisional reservation setting information, and outputting the identification information of the server selected after the provisional reservation setting is normally completed as the server selection result. Claims 1 to 9 characterized in that
A server selection device given in any 1 paragraph among them. 11. When the identification information of the user is input and whether or not the server selection device can select the server is calculated based on the position information of the user and the selection is determined, the server is selected. If the server selection processing by the selection means is continued and it is decided not to select,
And outputs the identification information of the candidate to become the server selection device for performing server selection for the user, any one of claims 1 to 10, characterized in that it comprises interrupting the server selection allowability determining means server selection process The server selection device described in 1. 12. A server selection method for selecting a server to be accessed by a user from a plurality of distributed servers, wherein the user identification information and a server list which is a list of the servers are input, Distance acquisition procedure for acquiring distances to the user for a plurality of servers
And a remaining resource amount acquisition procedure for acquiring the remaining resource amount of the server for the predetermined server, the user identification information, the server list, and the required server resource amount are input and included in the server list. Is
In addition, the remaining server resource amount acquired by the remaining resource amount acquisition procedure is determined by the distance acquisition procedure from the servers having the required server resource amount higher than the required server resource amount.
An output Surusa over server selection procedure the distance is short server with the user preferentially selected server identification information as the server selection result indicated by the input sought Te, the server, Yoo
User authentication function to limit user access
If there is, select it in the server selection procedure when selecting the server.
Server identification information, user identification information, and server
Enter the identification information of the service to the server.
Information for setting user authentication related to the
Generated based on the entered information (referred to as authentication setting information)
Then, the generated authentication setting information is output and the authentication setting is correct.
Always quit the selected server's identification information from the server
A server selection method comprising an authentication information notification procedure that is output as a selection result . 13. The distance acquisition procedure stores identification information of a relay device on a route from a target server selection device to a server for each of the plurality of servers, acquires identification information of the relay device, and acquires the identification information. 13. The server selection method according to claim 12 , further comprising a step of calculating a distance to the user for each of the plurality of servers based on the identification information of the relay device. 14. The distance acquisition procedure acquires position information of the user from identification information of the user, stores position information of the server in a server position table for each of the plurality of servers, and acquires the position information of the acquired user. 13. The server according to claim 12 , further comprising a step of obtaining a distance to the user for each of the servers based on identification information and position information of each of the servers stored in the server position table. Selection method. 15. The remaining resource amount acquisition procedure of claims 12 14, characterized in that it comprises the steps of storing the remaining resource information indicating the remaining resource amount for each of the plurality of servers in the server resource management table The server selection method according to any one of items. 16. The number of terminals that can be simultaneously connected to the server is defined as the maximum number of simultaneous connections, and the remaining resource amount is the number of terminals that are connected to the server from the maximum number of simultaneous connections of the server. 16. The server selection method according to claim 12, wherein the server selection method is a value obtained by subtracting. 17. The maximum total rate is defined as the sum of the rates of data transfer that can be executed by the server, and the remaining resource amount is calculated from the maximum total rate of the server to the rate of data transfer being executed by the server. server selecting method as claimed in any one of claims 12 16, characterized in that a value obtained by subtracting the sum of. 18. The sum total of the rates of data transfer to and from the outside of the sites that can be executed by the servers installed at the same site is defined as the maximum total rate of the sites, and the remaining resource amount is
Claims 12, characterized in that the base up to the total rate of the installation base of the server is a value obtained by subtracting the sum of the rate of data transfer between the base outside in the server group installation base of the server is running 15 The method for selecting a server according to any one of the items. 19. A remaining resource subtraction for subtracting, from the remaining resource amount stored in the server resource management table, the remaining resource amount of the server selected by the server selection method by the required resource input to the server selecting means. The server selection method according to claim 15 , further comprising a procedure. 20. The identification information of the service provided by the server and the server identification information selected by the server selection method are input, and the time required to provide the service (hereinafter referred to as the service provision time) is input based on the input. The calculated service provision time, the identification information of the server selected by the server selection method, and the required resource amount are input, and the server resource management is performed after the service provision time elapses from the time when these are input. 16. The server resource management timer procedure for updating the server resource management table so that the remaining server resource amount related to the server selected from the table information is increased by the required server resource amount. The server selection method described. 21. When a server to be managed by the server selection method has a temporary reservation function for temporarily securing a server resource for access by the user, the server is selected by the server selection method at the time of server selection. The server's identification information, user's identification information, service identification information are input, and information for making temporary reservation settings for the server (hereinafter referred to as temporary reservation setting information) is based on these information. A provisional reservation information notification procedure for generating and outputting the provisional reservation setting information and outputting the identification information of the server selected after the provisional reservation setting is normally completed as a server selection result is provided. The server selection method according to any one of Items 12 to 20 . 22. When the identification information of the user is input and whether or not the server selection device can select the server is calculated based on the position information of the user, and the selection is determined, the server is selected. If the server selection processing by the selection means is continued and it is decided not to select,
And outputs the identification information of the candidate to become the server selection device for performing server selection for the user, any one of the claims 12 to 21, further comprising a suspend server selection allowability determining steps server selection process Server selection method described in. 23. A computer-readable recording medium recording a server selection program for causing a computer to execute the processing procedure of the server selection method according to any one of claims 12 to 22 .