JP2008123334A - Distributed server system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a server system capable of optimizing a system capacity according to the number of users, and suppressing the deterioration of reliability for the users to the minimum. <P>SOLUTION: A distributed server system comprises: a plurality of servers at least including a data storage means and providing the same service to the user with a card type information medium; and a gateway for registering correspondence information, where each card type information medium is associated with either one server, and routing the server corresponding to the card type information medium when access is performed from the card type information medium. When the servers are extended, the gateway additionally registers the correspondence information between the server and the card type information medium to which the server provides the service. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a server system, and more particularly to a distributed server system.

  With the spread of the Internet, the form of providing services from a server via a network is increasing. In order to realize such a service, it is necessary to prepare a server that provides necessary functions and data, but such a server is prepared for each service to be provided, and one server provides the service. It is common to perform processing for all the users who receive it and store all data relating to that service.

  However, when only one server is prepared, if the traffic volume and data volume continue to increase as the number of users increases, it becomes difficult to ensure a constant response speed. If a server having sufficient performance with respect to the number of users can be used, a constant response can be obtained, but a high-performance server that performs a large amount of processing at high speed is expensive. In addition, it is difficult to accurately select a server having appropriate performance by accurately predicting the number of future users in the system design stage. Even if an expensive high-performance server is prepared in accordance with the number of users expected in the future, the server capacity cannot be fully utilized in the initial stage where the number of users is small, resulting in waste.

  In systems such as electronic money and POS systems, trust from users is important. One cause of damaging trust from users is the suspension of all services due to some cause. A system having one server as described above has low fault tolerance and may impair user reliability. In other words, if the only server fails to operate normally due to some trouble, the entire service must be stopped. A similar problem occurs when replacing a server with a new, higher performance server, such as when the number of users increases beyond the server's performance.

As a method for improving fault tolerance, it is conceivable to make servers redundant. For example, as shown in FIG. 4, by preparing a plurality of servers 60 having similar functions, the servers 60 are duplicated and connected to the network 30 by a gateway such as a load balancer 70 to control traffic. A system is conceivable. In this system, even if one server 60 ₁ goes down, the other server 60 ₂ having the same function continues to operate normally, so that the risk of service stoppage decreases. Improves fault tolerance.

  However, in such a duplex system, a plurality of servers 60 need to have similar data between the servers 60, so data verification must be performed between the servers 60. This data collation operation requires complicated processing, and when data collation fails, more complicated processing is required for recovery, so that the processing capability of the server for providing services is reduced. In addition, the duplex system must be a very expensive system because a plurality of expensive servers 60 must be prepared.

  The present invention has been made on the basis of the above circumstances, and makes it possible to optimize the system capacity in accordance with the number of users of the service, and the reliability seen from the user even when a part of the system is stopped. It is an object of the present invention to provide a server system capable of minimizing a decrease in performance.

  A distributed server system for providing a service to a user of the card type information medium based on the card type information medium of the present invention that achieves the above object includes at least data storage means, and the same service for the user of the card type information medium And a server corresponding to the card type information medium when the corresponding information corresponding to any one of the servers is registered and accessed from the card type information medium. A gateway for routing, and the gateway is configured to be able to additionally register correspondence information between the server and the card-type information medium provided by the server when the server is added. It is characterized by.

  In the above invention, the storage means included in the server may be a volatile memory.

  In the above invention, further comprising a shared database for backing up the data of the plurality of servers, wherein each of the servers relates to a card-type information medium stored in the storage means of each server to the shared database at a predetermined timing. , And the shared database may receive and store the data.

  Alternatively, a plurality of gateways may be provided, the plurality of gateways may have different network addresses, and access from a card-type information medium may be simultaneously performed on the plurality of gateways.

When multiple servers that provide the same service to users of card type information media and correspondence information that associates each card type information medium with one of the servers are registered and accessed from the card type information medium In addition, by using a gateway that routes to a server corresponding to the card-type information medium, each server needs to process only a part of users, so a system can be configured with an inexpensive server. At the same time, the risk of impairing the reliability of the user can be reduced.
Further, when the server is added, when the server is added, the correspondence information between the server and the card type information medium that the server provides a service can be additionally registered, thereby increasing the number of users. Accordingly, it is easy to add a server as appropriate, so that an appropriate system capacity can always be obtained.

[Embodiment 1]
A first embodiment of the present invention will be described below with reference to the drawings. FIG. 1 schematically shows a configuration for implementing a distributed server system according to the present invention. In this embodiment, the card-type information medium is a contactless IC card, but any card may be used as long as it includes information such as a credit card.

The distributed server system for providing a service to a user of the contactless IC card based on the contactless IC card in the present embodiment includes at least data storage means, and provides the same service to the user of the contactless IC card. When a plurality of servers 10 (10 ₁ ,..., 10 _n ) to be provided and correspondence information in which each non-contact IC card is associated with one of the servers are registered and accessed from the non-contact IC card And a gateway 20 for routing to a server corresponding to the contactless IC card.

The server 10 and the gateway 20 are connected to a network 30, and a non-contact IC card reader 40 (40 ₁ ,..., 40 _m ) as card type information medium reading means is connected to the network 30. The non-contact IC card reader 40 reads information stored in a non-contact IC card of a user who desires to provide a service and transmits the read information to the server 10. There can be a plurality of non-contact IC card readers 40. For example, it is assumed that each non-contact IC card reader 40 is installed in each store of a convenience store chain. Each user can receive the same service from any contactless IC card reader 40. By using the distributed server system including the server 10 and the gateway 20, the network 30, and the non-contact IC card reader 40, a service providing system that provides a service is constructed.

  In the present embodiment, the network 30 is the Internet. As a service provided by the service providing system according to the present invention, for example, an electronic money service using a non-contact IC card, a point service, and the like can be considered.

  In FIG. 1, both the server 10 and the gateway 20 are connected to the network 30, but may be configured in any way as long as they are connected to the non-contact IC card reader 40 through the network. For example, only the gateway 20 may be connected to the network 30 and the server 10 may be connected to the gateway 20.

  In the present embodiment, there are a plurality of servers 10 included in the system, and it is assumed that the number of servers 10 is increased according to the number of users. The gateway 20 is configured to be able to additionally register correspondence information between the server 10 and the contactless IC card that provides the service when the server 10 is added.

  In the present embodiment, first, a non-contact IC card possessed by a user registered for a service provided by the distributed server system is distributed and assigned to one of the servers 10. As a method for identifying the user, a card ID included in the non-contact IC card can be used.

The assignment of users to each server 10 is performed, for example, by determining the maximum number of users that can be assigned according to the processing capability of each server 10 and allocating the card ID to each server 10 within that capacity. If the total number of users is 10,000, for example, a server having the capability of associating a user ID of 0 to 9999 with a card ID possessed by each user and processing 1000 users with one server is provided. Prepare pieces. Then, the first server 10 ₁ assigned to the card ID to 999 from 0th user number (i.e. user), the second server 10 ₂ assigns the card ID of 1999 from No. 1000. By assigning up to the 10th server 10 _{10 in} the same procedure, 1000 different card IDs are assigned to each server 10.

  In the gateway 20, correspondence information indicating which card ID is assigned to which server 10 is registered. Various other methods such as a method of randomly assigning users to each server 10 can be considered. Each server 10 stores necessary information about the assigned card ID number in a storage unit provided in each server 10. Further, the number of people assigned to each server 10 may be determined individually based on the performance of each server 10, and the number of users assigned to each server 10 does not have to be the same. In the present embodiment, nonvolatile storage means such as a hard disk is used as storage means for each server.

  Next, what kind of processing is performed in the service providing system when the user desires a service will be described.

  For example, it is assumed that the card ID of a contactless IC card possessed by a user who desires a service is associated with the first user number. When the user brings the non-contact IC card close to the non-contact IC card reader 40, the card ID and other necessary information are read, and a service request is transmitted to the gateway 20 together with these information.

The gateway 20 that has received the information and the service request extracts the card ID from the information, and the server associated with the IC card ID is the first server 10 based on the registered card ID and server correspondence information. It found to be _1, to route the service request to the first server 10 _1. Server 10 ₁ that has received the service request executes the process corresponding to the service. The server 10 _1, and stores the information of the user as required, to update the pre-stored information.

  Here, the feature of the present invention is that each server is divided for each user, processing is performed only for the user associated with each server 10, and basically, other users are associated with each server 10. The server has no information and does no processing.

  For this reason, the amount of traffic to be processed by each server 10 is less than that of a server that processes all users at once. Therefore, even a relatively low-performance and inexpensive server can ensure a sufficient response speed, and can provide a service with the same quality as when an expensive high-performance server is used. Moreover, since the server 10 is divided for each user, even if the number of users increases, a server that can handle the increase is added, and the correspondence information between the user and the server for the increase is simply updated in the gateway. The overall capacity can be increased and unnecessary investment can be avoided.

  In the server system using this embodiment, if some servers go down, the service is stopped only for the users assigned to the servers, and the users assigned to the servers operating normally have no effect. It is possible to continue using the service without receiving it.

For example, in the service providing system utilizing the present embodiment, the server 10 ₁ and down for some reason. In this case, even if the user A assigned to the server 10 ₁ requests the provision of service from the IC card reader 40 _{1, the} user A cannot receive the service. If another user B who sees this (assuming that it is associated with the server 10 ₂ ) tries to request a service from the same IC card reader 40 ₁ , the service can be provided in this case. From the viewpoint of the user, it does not appear that the service providing system is malfunctioning. Therefore, it is possible to prevent a decrease in the reliability of the user with respect to the service providing system.

  In the above example, for example, when the service that the user A desires to provide is electronic money, the service request of the user A can be satisfied by the user B paying instead of the user A There is also. As described above, the situation in which only some users cannot use the service is much less serious than the situation in which all the services are down.

  In other words, according to the system of the present embodiment, it is stable while minimizing the reliability of the user at a very low cost compared to a server system having a redundant server that requires a plurality of expensive servers. Service can be provided.

[Embodiment 2]
A second embodiment of the present invention will be described with reference to FIG. 1 as in the first embodiment. The basic configuration is the same as that of the first embodiment. However, the storage unit of each server 10 which is a non-volatile storage unit in the first embodiment is different in that it is a volatile memory such as a DRAM in this embodiment. The other points are the same as in the first embodiment.

  In a system in which each server executes processing for all users, each server must store information on all users. Therefore, if volatile memory is used as storage means for the server, sufficient capacity is secured. It may not be possible. However, in this embodiment, since the number of users corresponding to each server can be determined in advance, this number of users may be set to a number that can be handled by the capacity of the volatile memory to be installed. In this way, each server does not need to access the hard disk every time there is a service request from the user, so that higher-speed processing can be realized. If the number of users increases, the server itself may be added.

[Embodiment 3]
A third embodiment of the present invention will be described with reference to FIG. Although the basic configuration of this embodiment is the same as that of the first or second embodiment, the present embodiment further includes a shared database 50 that backs up data of a plurality of servers 10. This embodiment is particularly effective when applied to the second embodiment in which the storage means of each server is a volatile memory. This is because volatile memory has a particularly high need for data backup because data is lost when the power is turned off.

  Each server 10 transmits data related to the card type information medium stored in the respective storage means to the shared database 50 at a predetermined timing, and the shared database 50 stores the received data.

  For example, the data transmission timing can be synchronized with the access timing of the server 10 from the user. In such synchronous data transmission, when there is frequent access from the user, data transmission frequently occurs, and there is a possibility that the processing capacity of the server 10 for providing the service is reduced. Alternatively, asynchronous data transmission, for example, a method of transmitting information updated at regular intervals can be considered. In addition, data transmission such as an event driven type may be used.

According to the present embodiment, even when a certain server 10 _x goes down, the shared database 50 stores information about the server 10 _x , so a new server 10 _y is prepared and the shared database 50 is stored. from the new server to 10 _y, by copying the information server 10 _x and which has been down has been stored, it is possible to recover the service immediately without losing the user information.

  Further, when it is desired to check access information to all the servers 10 and acquire statistical data, etc., if the shared database 50 is provided, the shared database 50 is accessed without accessing each server 10 individually. As a result, the data possessed by all the servers 10 can be obtained, and therefore, there is an advantage that the data can be obtained efficiently. Other points are the same as those in the first or second embodiment, and detailed description thereof is omitted.

[Embodiment 4]
A fourth embodiment of the present invention will be described with reference to FIG. The basic configuration is the same as that of the first to third embodiments. In this embodiment, a plurality of gateways 20 (20 ₁ , 20 ₂ ) are provided, and different network addresses are assigned to the gateways. Yes. The network address is, for example, an IP address when the network 30 is the Internet.

  In the distributed server system shown in the first to third embodiments, since a plurality of servers 10 are prepared, as described above, even when one server goes down, the entire service does not stop. However, since there is only one gateway 20, if the gateway 20 goes down due to some trouble, the entire service is stopped.

  On the other hand, in this embodiment, when a user who desires to use a service requests a service, the contactless IC card sends two service requests to the IP addresses of the two gateways 20. And send them separately separately at the same time.

In this case, since there are a plurality of gateways 20, if they operate simultaneously, the output signals from the gateways 20 are complicated and a problem arises. As a control method for the gateway 20 to solve such a problem, it is conceivable various methods, for example, one of the gateways 20 ₁ only by and running in an active state, other gateways 20 ₂ to the standby state Can be considered.

Then, when the active gateway 20 ₁ is down, to run by active gateways 20 ₂ in the standby state. Thereby, even when one gateway goes down, the service can be continuously provided.

The schematic block diagram of Embodiment 1 and Embodiment 2 of this invention is shown. It is a schematic block diagram of Embodiment 3 of this invention. It is a schematic block diagram of Embodiment 4 of this invention. It is a block diagram of the service provision system using a duplex server system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Server 20 ... Gateway 30 ... Network 40 ... IC card reader 50 ... Shared database 60 ... Server 70 ... Load balancer 80 ... Service request terminal

Claims (4)

A distributed server system that provides a service to a user of a card type information medium based on the card type information medium,
A plurality of servers including at least data storage means and providing the same service to users of the card-type information medium;
Corresponding information in which each card type information medium is associated with one of the servers is registered, and when accessed from the card type information medium, a gateway that routes to the server corresponding to the card type information medium,
The gateway is configured such that, when a server is added, correspondence information between the server and a card type information medium that provides the service can be additionally registered. .   2. The distributed server system according to claim 1, wherein the storage means included in the server is a volatile memory. The distributed server system according to claim 1 or 2,
And a shared database for backing up the data of the plurality of servers,
Each server transmits data related to the card-type information medium stored in the storage means of each server to the shared database at a predetermined timing;
The shared database receives and stores the data;
A distributed server system characterized by that.   4. The distributed server system according to claim 1, wherein a plurality of the gateways are provided, the plurality of gateways have different network addresses, and access from a card-type information medium is the plurality. A distributed server system, which is performed simultaneously on the gateways of the servers.

Images