JP2005071031A - Queuing system, and method and program for queue process - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently process a plurality of requests, making it possible for a server to process all the requests using its processing capability under normal operation even if it is temporarily under high load. <P>SOLUTION: In a request storage process, a classification changing part executes a classification changing process if the requirements for changing classification are met. On receiving a request, an allocator determines the classification of the request and determines where to allocate it and sends the request to the request queue of its destination. The request queue stores the request sent. In this way the request of the classification allocated in advance is stored in each request queue. That is, a group of requests stored in each request queue can be batch processed by each server. Thus, a common process that suits the plurality of requests stored in each request queue can be performed as a request process. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention executes a process for a plurality of requests for each classification in which a part or all of the processes for a request can be collectively executed by a common process, and can efficiently execute a process for a plurality of requests. The present invention relates to an apparatus, a queue processing method, and a queue processing program.

  Conventionally, in a server / client system, a request queue that is a queue of requests from clients to a server is stored, and requests are sequentially extracted from the request queue by a plurality of server processes or server computers, and processing is performed for each request. After that, it responds to the request source.

  In request response processing for responding to requests from clients as described above, a method of grouping request queues according to bandwidth, application attributes, or requesting IP address is proposed to guarantee the quality of service. (For example, refer to Patent Document 1).

  Also, in request reply processing, in order to process a plurality of requests at once, the same key in the search request is stored in the same area, and the processing for the search request stored in the same area is summarized. (For example, refer to Patent Document 2).

JP 2002-111721 A JP-A-4-123238 (Page 3, upper right column, line 16-lower right column, line 9, FIG. 2)

  By the way, in a system that executes request response processing, processing capacity at the server is required in proportion to the increase in requests, so in order to guarantee processing within a certain time for all requests, for example, temporarily It is necessary to configure and operate the server with the ability to handle all of them even in an overloaded state. However, in general, there is a possibility that an overload condition will not occur and it will not be a problem, and even if it becomes an overload condition, there are many times in time, so be prepared for overload. Is difficult in terms of system performance and cost.

  Therefore, when it becomes difficult to execute the request response processing for all requests, for example, as in Patent Document 1, by abandoning the processing of requests stored in a low priority queue. It corresponds.

  In Patent Document 2, a plurality of requests are processed in a lump. However, it can be dealt with only when there are a plurality of the same requests that can return the same search results as they are. In other words, when the response content is partially different or when it is necessary to change the response content according to the request timing, it is not possible to cope with the batch processing. Therefore, it cannot be said that the provision for overload is perfect.

  The present invention solves the above-described problems, can efficiently execute processing for responding to a plurality of requests, and can perform the same server processing as during normal operation even during a temporary high load. The goal is to be able to handle all requests with the ability.

  A queuing system according to the present invention is a queuing system (eg, queuing system 100) that queues a request received via a communication network and processes a request in the queue to reply to the request. Depending on the content of the received request, the request classification is specified from a plurality of classifications, and a request distribution unit (for example, distribution unit 21) that distributes the requests according to the specified classification is assigned to one of the plurality of classifications. Among the requests distributed by the request distribution means, a plurality of request queues (for example, request queues 22a, 22b, and 22c) that store requests of the assigned classification, and a plurality of requests that are stored in the request queue In Response processing execution means (for example, the servers 31, 32, and 33 shown in FIG. 1, or FIG. 4) that executes response processing (for example, steps S302 to S305) that generates response information for answering for each request queue. Server 31), and the plurality of classifications are classified according to request contents so that a plurality of requests executed by a common process (for example, step S302) belong to the same classification. The response processing execution means reads a plurality of requests stored in the request queue (for example, step S301), and part or all of the response processing for responding to the plurality of requests (for example, step S301). For example, according to the process of searching the stored contents of the databases 40, 41, 42, and the search result Processing for performing various determination, and performing collectively at least partially) the common processing of the various processes, such as processing of generating response information in accordance with the determination result.

  The response processing execution means determines whether or not individual processing for each request needs to be executed in the response processing (for example, step S303), and executes individual processing (for example, step S304) if necessary. Response information for each request may be generated in order to respond for each request.

  In response to the establishment of a predetermined classification change condition (for example, step S201), a classification setting changing unit (for example, a classification changing unit 23) that changes the setting contents of a plurality of classifications is provided. The setting content may be changed by executing a process of changing one of the request queues to each of the changed classifications (step S202).

  The classification change condition is established when the number of requests received per unit time exceeds a predetermined number (for example, 10,000), and the classification setting changing means sets the number of classifications when the classification change condition is established. You may make it perform the process which reduces (for example, reduces from 3 classifications to 2 classifications), and expands the range of the request which belongs to each classification.

  The classification change condition is satisfied when there is a predetermined change instruction (for example, a classification change instruction by a system administrator using an input device included in the queuing device 20). May be configured to change.

  A plurality of response processing execution means (for example, servers 31, 32, 33) are provided, and each of the plurality of response processing execution means is assigned to one of the request queues (for example, each of the servers 31, 32, 33 is a request queue). 22a, 22b, and 22c), a response process is executed to reply to a plurality of requests stored in the assigned request queue, and when the classification changing unit satisfies a predetermined classification changing condition, You may make it change the allocation of the request queue in each of several response process execution means.

  The request includes position information for requesting a surrounding map, and the plurality of classifications are set in advance by being classified for each position range indicated by the position information, and the response processing execution means stores the map information. The common map database may be searched and a common process may be executed to generate a common map including each neighboring map corresponding to a plurality of requests stored in the request queue. That is, the queuing system may be the map information system 100a for providing a surrounding map in response to a request.

  The request includes reservation request information for requesting a ticket reservation. Each of the plurality of classifications is set in advance for each event that accepts at least a ticket reservation. Search the ticket database that stores reservation status information, determine whether or not to accept a reservation for some or all of the requests in response to multiple requests stored in the request queue, and depending on the determination result Thus, a common process for generating a common reservation result message for a part or all of the request may be executed. That is, the queuing system may be a ticket reservation system 100b for accepting a ticket reservation in response to a request.

  The request includes correction request information for requesting correction of the answer, and a plurality of classifications are set in advance for each test question, and the response processing execution means stores the answer information of the test question. The ticket database is searched, and it is determined whether or not some or all of the requests are correct according to a plurality of requests stored in the request queue, and one of the requests is determined according to the determination result. It may be configured to execute a common process for generating a common correction result message for some or all. In other words, the queuing system may be a test correction system (test scoring system 100c) for determining whether or not an answer to a test question is correct in response to a request.

  In addition, the queuing device of the present invention queues a request received via a communication network, and distributes and stores the requests in the queue for each predetermined classification in order to reply to the request (for example, A queuing device 20) that identifies a request classification from a plurality of classifications according to the content of the received request and distributes the requests according to the identified classification, and assigns one of the plurality of classifications Among the requests distributed by the request distribution means, and a plurality of request queues each storing a request of the assigned classification, and each of the plurality of request queues is stored in the request queue. Requests that are made over a communications network The response processing execution device that executes response processing for generating response information for responding to a plurality of requests stored in the request queue is generated by generating response information for each request queue. In order to execute a part or all of them in a batch by a common process, a request for the assigned classification is stored, and the plurality of classifications used by the request distribution means for distribution are respectively a part of the response process or It is characterized in that a plurality of requests that are all executed by a common process are classified and set in advance so as to belong to the same classification.

  A classification setting changing unit that changes setting contents of a plurality of classifications in response to the establishment of a predetermined classification changing condition, and the classification setting changing unit executes a process of changing the contents of the classification, and The setting content may be changed by executing a process of assigning one of the request queues to each of the plurality of classifications.

  The classification change condition is satisfied when the number of requests received per unit time exceeds a predetermined number, and the classification setting change means reduces the number of classifications to each classification when the classification change condition is satisfied. It may be configured to execute processing for expanding the range of the request to which it belongs.

  The classification change condition may be satisfied when a predetermined change instruction is given, and the classification setting change unit may be configured to change the setting contents according to the change instruction.

  The response information receiving means for receiving the response information generated for each request queue by the response processing execution device via the communication network, the received response information, and the response content uniquely determined by the response information When the information is recognized (for example, when the reservation reception has ended), information indicating that (for example, information indicating that the reservation reception has ended) and response information storage means for storing the information in association with the request queue If the response content to the request stored in the predetermined request queue is uniquely determined (for example, when information indicating that the reservation reception is completed is stored), the response processing execution device The request for each request stored in the predetermined request queue Towards strike source reply information stored in the response information storage means (e.g., a common message that can not be reserved) may include a response information transmitting means for transmitting through the communication network.

  The queuing device of the present invention is a queuing device that queues requests received via a communication network, and distributes and stores the requests in the queue for each predetermined classification in order to respond to the request. Depending on the content of the received request, the request classification means that identifies the classification of the request from the plurality of classifications and distributes the requests according to the identified classification, and one of the plurality of classifications is assigned. Among the requests distributed by the means, a plurality of request queues each storing a request of the assigned classification, and a predetermined number (for example, one or two) of the requests stored in the request queue are communicated Read through the network and read the predetermined number A response information receiving means for receiving, via a communication network, response information generated for each predetermined number of requests from a response processing execution device for executing response processing for generating response information for responding to the received request; Response information transmitting means for transmitting the received response information to the request source of the corresponding request via a communication network, and the plurality of classifications used by the request distribution means for distribution are each a response processing execution device Are set in advance so that a plurality of requests that can be used in common by the response information belong to the same category, and the response information transmitting means indicates the received response information. If the response content is a predetermined content (for example, a message indicating that reservation is not possible), the corresponding request key The same response information as the received response information (for example, a message indicating that the reservation cannot be made) is sent to the request source of each request stored in-and the request stored in the request queue is deleted. The process which performs is performed.

  Further, the queue processing method of the present invention is a queue processing method for queuing a request received via a communication network and processing a request in the queue in order to reply to the request. Accordingly, a step of identifying the request classification from a plurality of classifications, a step of distributing requests according to the identified classification, and a plurality of classifications are assigned, and among the requests distributed by the request distribution means, A step of storing the assigned classification request in each of the plurality of request queues, a step of reading out the plurality of requests stored in the request queue, and response information for replying to the plurality of read requests. Response generated every time And performing a part or all of the processing in a batch by a common process, and the plurality of classifications are such that a plurality of requests executed by the common processing are partly or entirely part of the reply process belong to the same class. , And is preset according to the contents of the request.

  In response to the establishment of a predetermined classification change condition, a process for changing the contents of the classification is executed, and a process for assigning one of the request queues to each of the plurality of changed classifications is executed, whereby a plurality of classifications are executed. It may be configured to include a step of changing the setting contents.

  Further, the queue processing program of the present invention is a queue processing program for queuing a request received via a communication network and processing a request in the queue in order to respond to the request. In the ingress device 20, the servers 31, 32, and 33), the step of identifying the classification of the request from a plurality of classifications according to the content of the received request, the step of distributing the requests according to the identified classification, Among the requests distributed by the request distribution means, a step of storing the allocated classification requests in a plurality of request queues respectively, and reading out the plurality of requests stored in the request queue Step And executing a step of performing a part or all of the response processing for generating a response information for responding to the plurality of read requests for each request queue in a batch by the common processing, and each of the plurality of classifications is a response. It is characterized in that the request is classified and set in advance so that a plurality of requests executed by a common process are partly or entirely part of the common process.

  In addition to executing a process for changing the contents of the classification and a process for assigning one of the request queues to each of the plurality of changed classifications according to the fact that a predetermined classification changing condition is satisfied on the computer. Thus, the step of changing the setting contents of the plurality of classifications may be executed.

  According to the present invention, a configuration is set in which a category having a possibility of being processed in a batch by common processing is set in advance, and a plurality of requests belonging to the category are collectively stored in a request queue provided for the category. As a result, part or all of the processing for responding to multiple requests stored in one request queue can be processed simultaneously in one common process, and responses to multiple requests can be made simultaneously. Can efficiently execute the processing to perform server processing, reduce the processing load on the server, and improve the response capability at the time of temporary high load while maintaining the same server capability as during normal operation Can do. Therefore, it is possible to improve the degree of expectation that responses to all requests are completed even during a temporary high load.

  In addition, when the load on the server increases, the load on each server can be reduced by performing a classification change that reduces the number of classifications and expanding the range to be handled by common processing. Therefore, even if the load temporarily increases, the processing corresponding to the request can be executed without improving the processing capability of the server.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram illustrating an example of a configuration of a queuing system 100 including a queuing device 20 according to an embodiment of the present invention. The queuing system 100 includes a client computer (hereinafter referred to as “client”) 10, a queuing apparatus 20, and a plurality of server computers (hereinafter simply referred to as “servers”) 31, 32, 33.

  The client 10 and the queuing device 20 are connected to a communication network such as the Internet. In addition, the queuing device 20 and each of the servers 31, 32, and 33 are connected by a communication network such as a dedicated line. Any number of clients 10 may be provided.

  The client 10 is configured by an information processing device such as a personal computer, for example, and is managed by a user of the system 100. The user terminal 40 has various functions such as a connection function for connecting to a communication network and a browser function for displaying a markup language such as HTML (HyperText Markup Language).

  The queuing device 20 is configured by an information processing device such as a WWW (World Wide Web) server. The queuing device 20 has various functions such as a function for distributing and holding requests transmitted from the client 10 via the communication network and classifying the requests according to the content of the request.

  The request includes user information for specifying the client and information indicating the content of the request. The information indicating the content of the request includes information serving as a key for determining the classification (for example, when requesting a surrounding map, position information indicating the center position of the surrounding map (see Example 1)). .

  The queuing device 20 includes a distributor 21, a plurality of request queues 22a, 22b, and 22c, and a classification changing unit 23. Each request queue 22a, 22b, 22c is assigned a “classification” in advance. Note that the queuing device 20 includes a control unit (not shown) that controls the entire device 20, and this control unit executes processing according to a control program stored in a storage medium (for example, ROM) included in the device 20. In this example, it is assumed that the control means is mounted on the sorter 21 and the classification changing unit 23, respectively.

  The distributor 21 distributes the request transmitted from the client 10 to each request queue 22a, 22b, 22c according to the content of the request and the classification assigned to each request queue 22a, 22b, 22c. Has the function to execute.

  Each of the request queues 22a, 22b, and 22c is configured by a rewritable storage medium such as a RAM. In each of the request queues 22a, 22b, and 22c, a request that conforms to a previously assigned classification is queued (meaning that it is stored as a request queue).

  The “classification” assigned to each of the request queues 22a, 22b, and 22c is, for example, a classification classified according to the range of request contents. Specifically, it is determined in advance that each request that may be handled is assigned to one of the three classifications of requests A, B, and C according to the contents thereof. In this case, the requests A, B, and C are classified for each range of request contents that each server 31, 32, and 33 can process in a batch by common processing. Each predetermined request classification (request A, B, C) is assigned to one of the request queues 22a, 22b, 22c.

  The classification changing unit 23 has a function of arbitrarily changing the classification of requests assigned to the request queues 22a, 22b, and 22c. For example, the classification changing unit 23 executes a process of changing the number of request classifications, changing the range of each classification, and determining the correspondence between the changed classification and each request queue 22a, 22b, 22c. . Furthermore, when the classification is changed, the classification changing unit 23 performs processing according to the requests stored in the request queues 22a, 22b, and 22c and the request queues 22a, 22b, and 22c. , 33 is determined, and the server 31, 32, 33 is notified of the determination result.

  The classification changing unit 23 has a function of automatically processing a classification change. For example, the classification changing unit 23 is used as an interface for a system administrator to arbitrarily instruct a classification change from the outside of the queuing device 20. It may be. If comprised in this way, according to the classification change instruction | indication by a system administrator, a classification change can be performed and system management can be performed smoothly.

  Each of the servers 31, 32, and 33 is configured by an information processing device such as a WWW server, for example. Each of the servers 31, 32, and 33 is a server device that executes various processes such as a function of performing various processes in response to a request from the client 10 input via the queuing device 20.

  Each of the servers 31, 32, and 33 has a function of reading out requests stored in the request queues 22a, 22b, and 22c, that is, requests in the queue, and executing processing according to the requests.

  Note that the queuing device 20 and each of the servers 31, 32, and 33 are managed by a provider of a predetermined service (for example, a search service such as a map or a restaurant) provided by using each of the servers 31, 32, and 33. The

  Next, the operation of the queuing system 100 according to one embodiment of the present invention will be described with reference to the drawings.

  First, processing until a request is distributed by the queuing device 20 and stored in the request queues 22a, 22b, and 22c will be described. FIG. 2 is a flowchart illustrating an example of a request storage process for sorting and storing requests in the queuing device 20.

  In the queuing apparatus 20, as the initial setting of the classification, the classification of the request and the correspondence between each classification and each request queue 22a, 22b, 22c are set in advance. Here, as an initial setting, three classifications A, B, and C corresponding to the contents of the request are determined, and the request queues 22a, 22b, and 22c are associated with the classifications A, B, and C, respectively. To do. Note that the initial setting may be automatically set when the queuing device 20 is activated, for example, and is set by a system administrator operating an input device (for example, a keyboard or a mouse) included in the device 20 in advance. You may do it.

  In the request storing process, the classification changing unit 23 of the queuing device 20 checks whether or not the classification changing condition is satisfied (step S201), and if it is satisfied, executes the classification changing process according to the content of the satisfied condition. (Step S202).

  The classification change condition may be determined by the type of server application, or may be determined by the number of servers, the number of requests processed, the number of requests stored in the current request queue, and the like. Further, the classification change process may be executed according to an instruction from the server administrator, or may be automatically executed according to the current operation status of the server.

  In this example, the classification change condition includes, for example, a condition (classification change condition X) that is established when the number of received requests per unit time exceeds a predetermined number (for example, 10,000), or a received request per unit time. It is assumed that there are a plurality of conditions such as a condition (classification change condition Y) that is satisfied when the number becomes a predetermined number (for example, 5,000) or less. The classification change process is automatically executed when the classification change condition is satisfied.

  In the classification change process in step S202, processing for changing the number of classifications and the correspondence between each new classification and each request queue 22a, 22b, 22c is executed according to the contents of the established classification change condition. The In addition, the classification changing unit 23 performs the processing corresponding to the requests stored in the request queues 22a, 22b, and 22c and the request queues 22a, 22b, and 22c in the classification changing process. , 33 is also determined. The determination result is notified to the necessary servers 31, 32, and 33 by the classification changing unit 23, respectively.

  Specifically, when the classification change condition X is satisfied, the load on each of the servers 31, 32, and 33 is large, so the number of classifications is changed from, for example, three classifications (classifications A, B, and C) to two classifications (classifications). The request queues 22a and 22b are changed to correspond to the classifications a and b. In this case, the request queue 22c is not used. For example, the server 33 may perform processing corresponding to the request stored in the request queue 22b in a shared manner with the server 32. For example, the server 33 performs processing corresponding to the request stored in the request queue 22a with the server 31 and also performs processing corresponding to the request stored in the request queue 22b with the server 32. May be performed.

  If the classification is changed as described above, the range to be dealt with by the common processing when the load increases can be widened, and the load on each server 31, 32, 33 can be reduced. Therefore, the processing corresponding to the request can be executed without improving the processing capability of each of the servers 31, 32, and 33.

  Specifically, when the classification change condition Y is satisfied, the load on each of the servers 31, 32, and 33 is small, so the number of classifications is changed from, for example, three classifications (classifications A, B, and C) to four classifications. The request queues 22a, 22b, 22c, and 22d (request queue 22d is not shown) are changed to correspond to the categories x, y, z, and v. For example, the server 33 may sequentially perform processing corresponding to the requests stored in the request queue 22c and the request queue 22d. Further, for example, the processing corresponding to the request stored in the request queue 22d may be shared by at least two of the servers 31, 32, and 33.

  If the classification is changed as described above, the range to be dealt with by the common processing when the load is reduced can be narrowed, and the processing capability of each of the servers 31, 32, and 33 can be fully exhibited and the client can be reduced. Reply to the computer 10 can be made early.

  Next, when a request from the client 10 is received (step S203), the distributor 21 of the queuing device 20 determines to which class the request belongs according to the content of the received request, and according to the determination result. A distribution destination is determined, and a request is transmitted to the request queues 22a, 22b, and 22c of the distribution destination (step S204).

  Then, the request queues 22a, 22b, and 22c store the transmitted requests (step S205). Therefore, the assigned classification requests are stored in the request queues 22a, 22b, and 22c, respectively. That is, the request groups stored in the request queues 22a, 22b, and 22c have a possibility that the servers 31, 32, and 33 can be collectively processed, respectively.

  As described above, requests from each client such as the client 10 are sorted into any classification that may be processed in a batch and stored in the request queues 22a, 22b, and 22c. That is, each request queue 22a, 22b, and 22c stores a group of requests that can be collectively processed.

  Next, after a request from each client is distributed and stored in each request queue 22a, 22b, 22c, each server 31, 32, 33 executes processing according to each request, and makes a response to each request. Processing will be described.

  FIG. 3 is a flowchart illustrating an example of a request handling process executed by each server 31, 32, 33 in the queuing system 100. Here, it is assumed that requests from the client 10 are already stored in the request queues 22a, 22b, and 22c. Here, an example will be described in which the server 31 executes processing according to each request stored in the corresponding request queue 22a.

  The request handling process is executed at regular intervals in each of the servers 31, 32, and 33, for example. The predetermined time is determined in advance according to the content of the request to be handled, for example, whether it is necessary to respond early. Specifically, the fixed time is determined, for example, every second, every minute, every hour, or the like. The request handling process may be executed when notified from the queuing device 20 that the request queue is stored in each of the request queues 22a, 22b, and 22c, not limited to a certain time.

  In the request handling process, the server 31 reads all requests stored in the request queue 22a (step S301), and executes a common process for all requests (step S302).

  The common process is a process that is collectively executed so as to be able to respond to each of a plurality of requests, and means a common process that covers each process for a plurality of requests.

  In the common processing, there are a case where all requests are processed in batch and a case where a plurality of requests configured by a part of all requests are processed in batch. In the latter case, the common process is executed a plurality of times until all requests are covered in step S302.

  When the common process is finished, the server 31 confirms whether the individual process is necessary (step S303). That is, it is confirmed whether or not responses to all requests can be made only by common processing. For example, if all requests can be answered according to the result of common processing, it is not necessary to perform individual processing. If the result of common processing cannot reply to all requests (for example, response) If the content of the information to be executed is insufficient), it is necessary to perform individual processing.

  If it is determined that the individual processing is necessary, the server 31 executes the individual processing for each request that requires the individual processing (step S304).

  When the creation of the response information indicating the response to each request is completed, the server 31 transmits the response information to each client such as the client 10 that is the request source via the communication network (step S305). This processing result may be a result common to all requests or may be an individual result for a part or all of all requests.

  As described above, in the above-described embodiment, the requests from the client are assigned to predetermined classifications, stored in the request queue provided for each classification, and all the requests stored in each request queue are stored. Alternatively, since a part is processed in batch, the processing load of each server corresponding to the request can be reduced.

  In addition, as described above, a category having a possibility of being processed in a batch by common processing is set in advance, and a plurality of requests belonging to the category are collectively stored in a request queue provided corresponding to the category. Because it is configured, part or all of the processing for responding to multiple requests stored in one request queue can be processed simultaneously in one common processing, and the processing load of each server Can be mitigated, and the response capability at the time of temporary high load can be improved while maintaining the same server capability as during normal operation. Therefore, it is possible to improve the degree of expectation that responses to all requests are completed even during a temporary high load.

  In addition, as described above, when the load on the servers 31, 32, and 33 is increased, the classification is changed so that the number of classifications is decreased, and the range to be dealt with by the common process is expanded. The load on the servers 31, 32, and 33 can be reduced. Therefore, even when the load temporarily increases, the processing corresponding to the request can be executed without improving the processing capability of each server 31, 32, 33.

  In the above-described embodiment, whether or not the classification change condition is satisfied is monitored, and the classification change process is performed when the condition is satisfied. However, the classification is not changed and the initial setting is fixed. It is good also as a structure to perform (for example, the structure which does not perform step S201, S202). In this case, the classification changing unit 23 executes a process of changing the initial setting of the contents of the classification based on an instruction to change the initial setting by the system administrator.

  In the above-described embodiment, the three servers 31, 32, and 33 are provided. However, a process corresponding to each request stored in all the request queues may be performed by one server. In this case, for example, as shown in FIG. 4, the processing corresponding to each request group stored in the request queues 22 a, 22 b, and 33 c may be configured to be sequentially executed by one server 31. Even with such a configuration, the same effect as described above can be obtained.

  In the above-described embodiment, the queuing device 20 and the servers 31, 32, and 33 are configured as separate information processing devices, but the queuing device 20 and the servers 31, 32, and 33 are the same. The information processing apparatus may be used.

  In addition, the queuing device 20 and the servers 31, 32, and 33 in the above-described embodiment cause the queuing device 20 and the servers 31, 32, and 33 to queue requests received via the communication network, A queue processing program for processing the requests in the queue is provided in order to respond to the request. For example, the queue processing program specifies the classification of the request from a plurality of classifications according to the content of the received request to the queuing apparatus 20 or each of the servers 31, 32, and 33 (step S204), and the identified classification A request is distributed according to the step (step S205) and one of a plurality of classifications is assigned, and among the requests distributed by the request distribution means, requests of the assigned classification are assigned to a plurality of request queues 22a and 22b. , 22c (step S205), reading a plurality of requests stored in the request queues 22a, 22b, 22c (step S301), and reply information for replying to the read requests. Riku Some of response processing for generating each Sutokyu or control program for executing and performing collectively by a common processing whole (step S302). The plurality of classifications are set in advance so as to be classified for each request content so that a plurality of requests executed by a common process in some or all of the response processes belong to the same classification.

  In addition, the queue processing program further executes a process for changing the contents of the classification in response to the establishment of a predetermined classification change condition in the queuing device 20, and requests queues for each of the plurality of changed classifications. This is a control program that executes a step (for example, steps S201 and S202) for changing setting contents of a plurality of classifications by executing a process of assigning any one of the above.

  Hereinafter, an example in which the queuing system 100 according to the embodiment of the present invention described above is applied to a map information system 100a for queuing a browsing request for map information will be described as a first example.

  FIG. 5 is a block diagram illustrating an example of the configuration of the map information system 100a. The map information system 100a includes a client 10 (not shown in FIG. 5), a queuing device 20, servers 31, 32, and 33, and a map information database 40.

  The map information database 40 is composed of, for example, a large-capacity storage medium such as a hard disk, and stores vector data for displaying a map and various data such as symbols and figures displayed on the map.

  In this example, it is assumed that the request queues 22a, 22b, and 22c are classified corresponding to certain regional blocks.

  In this example, each server 31, 32, 33 reads vector data from the map information database 40 in accordance with the request stored in the corresponding request queue 22a, 22b, 22c, respectively. It has a function to create and reply. The two-dimensional planar map is created in an image format such as GIF (Graphics Interchange Format).

  Next, request storage processing in this example will be described. In the request storing process, when the class changing unit 23 of the queuing device 20 receives a request from the client 10 (see step S203), it determines the class according to the content of the request (see step S204).

  In this example, the request includes, for example, user information for specifying the address of the client 10 and location information for specifying a location where a surrounding map is desired. As the location information, for example, information indicating latitude and longitude, or a location name or information indicating a location is used.

  Further, the classification of the request is determined by, for example, the classification of the area block including the point indicated by the location information in the request (for example, the point specified by the latitude and longitude). That is, the classification assigned to the regional block including the point indicated by the location information in the received request is the classification of the request.

  Specifically, for example, as shown in FIG. 6, the regional blocks are classified into three blocks ranging from block A to block C, and classifications A to C are assigned to the respective blocks A to C. If the location indicated by the received request is included in the range of the block A, the classification A allocated to the block A is set as the classification of the request.

  In addition, the area blocks A to C are within a range in which the point indicated by the location information included in the request for providing the surrounding map can be specified. In this example, since a surrounding map of a predetermined range centering on the point indicated by the location information included in the request is provided, the range of the regional blocks A to C is created according to the request as shown in FIG. The range is narrower than the range of the map. Note that the map created in response to the request only needs to include the point indicated by the location information included in the request, and may not be a map centered on the point indicated by the location information included in the request.

  Next, the distributor 21 determines a request queue (for example, the request queue 22a) assigned in correspondence with the determined classification A as a distribution destination request queue, and transmits the request toward the request queue 22a (step S204). reference).

  Then, the request queue 22a receives the transmitted request and stores the received request (see step S205).

  As described above, each request of the classification assigned in advance is stored in each of the request queues 22a, 22b, and 22c. For example, when each classification setting is made such that the block A and the request queue 22a are classified as A, the block B and the request queue 22b are classified as B, and the block C and the request queue 22c are classified as C. When a plurality of requests are sent from each client, a plurality of requests for requesting a map around a predetermined point in the block A are stored in the request queue 22a, and the periphery of the predetermined point in the block B is stored in the request queue 22b. A plurality of requests for requesting a map are stored, and a plurality of requests for requesting a map around a predetermined point in the block C is stored in the request queue 22c.

  Next, request response processing in this example will be described. Here, a case where the server 31 executes a process according to a request stored in the corresponding request queue 22a will be described as an example.

  In the request handling process, the server 31 reads all requests stored in the request queue 22a (see step S301) and executes a common process for all requests (see step S302). In this example, in the common process, a process for collectively generating a peripheral map of the block A including each point indicated by the map information of each request stored in the request queue 22a is performed. Specifically, as shown in FIG. 7A, a common map including a map around each point (map Aa, map Ab, map Ac, map Ad, map Ae) indicated by the location information of each of the five requests. Generate.

  Next, the server 31 checks whether or not individual processing is necessary (see step S303). In this example, since the surrounding map created by the common processing does not match the surrounding map for replying to each request, the server 31 determines that individual processing is necessary. When a common map as shown in FIG. 7A is commonly used as response information for responding to each request, the server 31 determines that individual processing is not necessary. It is assumed that whether to respond using a common map or using an individual map created by individual processing is predetermined.

  If it is determined that the individual process is necessary, the server 31 executes the individual process for creating an individual map corresponding to each request (see step S304).

  In the individual processing, the server 31 cuts out the map area of the surrounding map corresponding to each request from the common map, and creates an individual map used as response information for responding to each request. For example, as shown in FIG. 7B, the individual map is a map created by cutting out a part of the common map, and is a surrounding map of each point indicated by the location information of each request.

  When the individual map that is the response information is created, the server 31 performs a process of transmitting the response information via the communication network to each client that is the request source (see step S305).

  As described above, in this example, a request for a map request from a client is stored in a separate request queue according to a predetermined classification for each region block, and is common to each request stored in each request queue. Since the configuration is such that the batch processing for creating the map is performed, it is possible to reduce the processing load of each server corresponding to the request for the map request.

  In other words, after creating a common map by common processing, it is configured to create an individual map using the common map. Compared to creating an individual map according to each request without creating a common map, The load on the server due to the process of reading the map data from the map information database 40 and the load on the server due to the process of creating the individual map can be reduced.

  In addition, as described above, a category having a possibility of being processed in a batch by common processing is set in advance, and a plurality of requests belonging to the category are collectively stored in a request queue provided corresponding to the category. Since it is configured, part or all of the processing for responding to multiple requests stored in one request queue can be processed simultaneously in one common processing, and the processing load on the server is reduced. It is possible to reduce this, and it is possible to improve the response capability at the time of temporary high load while maintaining the same server capability as during normal operation. Therefore, it is possible to improve the degree of expectation that responses to all requests are completed even during a temporary high load.

  Hereinafter, as a second example, an example in which the above-described queuing system 100 according to an embodiment of the present invention is applied to a ticket reservation system 100b for receiving and managing ticket reservation will be described.

  FIG. 10 is a block diagram showing an example of the configuration of the ticket reservation system 100b. The ticket reservation system 100b includes a client 10 (not shown in FIG. 10), a queuing device 20, servers 31, 32, and 33, and a ticket database 41.

  The ticket database 41 is configured by a large-capacity storage medium such as a hard disk, for example, and stores various data such as information indicating the ticket reservation status.

  In this example, it is assumed that the request queues 22a, 22b, and 22c are classified for each event that accepts reservations for tickets such as concerts and performances. For simplicity of explanation, it is assumed that each event does not have a seat grade (for example, S seat, A seat, etc.), and a reservation request is made in units of one seat.

  Also, in this example, each server 31, 32, 33 reads ticket information from the ticket database 41 according to the request stored in the corresponding request queue 22a, 22b, 22c, and is there a vacant seat? It has a function of confirming whether or not, deciding a seat for accepting a reservation from vacant seats, creating reservation acceptance information indicating a seat for which a reservation has been accepted, and responding.

  Next, request storage processing in this example will be described. In the request storing process, when the class changing unit 23 of the queuing device 20 receives a request from the client 10 (see step S203), it determines the class according to the content of the request (see step S204).

  In this example, the request includes, for example, user information for specifying the address of the client 10 and event type information for specifying an event for which a reservation is desired. As the event type information, for example, an identification number assigned in advance or information indicating an event name is used.

  Further, the classification of the request is determined by, for example, the classification of the event indicated by the event type information in the request. That is, the classification assigned to the event indicated by the event classification information in the received request is the classification of the request.

  Specifically, when the classifications A to C are assigned to the respective events A to C that are subject to the ticket reservation reception service, and the event indicated by the received request is the event A, the event The classification A assigned to A is the request classification.

  Next, the distributor 21 determines a request queue (for example, the request queue 22a) assigned to the determined classification A as a distribution destination request queue, and issues a ticket reservation request for the event A toward the request queue 22a. The request containing is transmitted (refer step S204).

  Then, the request queue 22a receives the transmitted request and stores the received request (see step S205).

  As described above, each request of the classification assigned in advance is stored in each of the request queues 22a, 22b, and 22c. For example, when each classification setting is made such that event A and request queue 22a are classified as A, event B and request queue 22b are classified as B, and event C and request queue 22c are classified as C. When a plurality of requests are sent from each client, a plurality of requests for requesting ticket reservation for event A are stored in the request queue 22a, and a plurality of requests for requesting ticket reservation for event B are stored in the request queue 22b. The request is stored, and a plurality of requests for requesting ticket reservation for the event C are stored in the request queue 22c.

  Next, request response processing in this example will be described. Here, a case where the server 31 executes processing according to a request stored in the corresponding request queue 22a will be mainly described as an example.

  Also, here, each classification setting is made such that the event A and the request queue 22a are classified as A, the event B and the request queue 22b are classified as B, and the event C and the request queue 22c are classified as C. Shall.

  FIG. 9 is an explanatory diagram showing a part of information stored in the ticket database 41. As shown in FIG. 9, the ticket database 41 stores reservation status information indicating the current reservation status for each event. Here, as shown in FIG. 9, in Event A, all seats are reserved for 2000 people, and currently 1900 seats are reserved, and in Event B, all seats are currently reserved for 5000 people. Is reserved, and in event C, the total number of seats is 10,000 and seats for 8500 are currently reserved.

  In the request handling process, the server 31 reads all requests stored in the request queue 22a (see step S301) and executes a common process for all requests (see step S302). In this example, in the common process, the seat reservation for the event indicated by each request stored in the request queue 22a is processed in a batch, and a process for creating a response message indicating whether the reservation is possible is executed in a batch. .

  Specifically, the server 31 confirms the number of requests stored in the request queue 22a and determines how many seats should be reserved. Next, the ticket database 41 is inquired about the number of remaining seats of the event A.

  If the number of remaining seats exceeds the required number as a result of the inquiry, the number of seats to be newly reserved is subtracted from the number of remaining seats of event A in the ticket database 41 and information regarding all the reservation persons is stored in the ticket database 41.

  Specifically, as shown in FIG. 9, since the number of remaining seats for event A is 100, the server 31 has a number of requests stored in the request queue 22a (that is, the number of reservation requests) of 100 or less. If so, the number of requests is added to the reserved number in the reservation status information stored in the ticket database 41, and the number of requests is subtracted from the number of remaining seats. Then, information for specifying the reservation person who has newly accepted the reservation is stored in the ticket database 41.

  As a result of the inquiry, there is no remaining number of seats more than the necessary number, but if there is a remaining number of seats, a person who can make a reservation is determined in the order of arrival of the request message, and a new number of remaining seats for event A in the ticket database 41 The number of seats to be subtracted is subtracted, and information relating to the person who can make a reservation is stored in the ticket database 41.

  Specifically, as shown in FIG. 9, since the number of remaining seats for event A is 100, the server 31 has a number of requests stored in the request queue 22a (that is, the number of reservation requests) of 101 or more. If so, among the reservation status information stored in the ticket database 41, a reservationable number of 100 is added to the reserved number and a reservationable number of 100 is subtracted from the remaining seat number. Then, information for specifying the reservation person who has newly accepted the reservation is stored in the ticket database 41.

  If there is no remaining number of seats as a result of the inquiry, the ticket database 41 is not updated.

  Next, in the common process (see step S302), the server 31 receives a reservation message for each user who has made a reservation (for example, “Event A ticket reservation accepted”). Is created for each user who could not make a reservation (for example, “there were no remaining tickets for Event A”).

  If the individual process is necessary, the server 31 executes the individual process (see steps S303 and S304). For example, if the system sends a message to the clients saying that “Mr. XX, there were no remaining tickets” as a message that the ticket cannot be reserved, the part “There were no remaining tickets.” Is created by common processing. Then, the individual name “Mr. XXX” is executed by individual processing. Therefore, when the response information including the individual message as described above is to be created, it is determined that individual processing is necessary. Here, in the individual processing, for example, individual name is extracted using user identification information called Cookie, and processing for completing reply information to each client is executed.

  When a message as response information is created, the server 31 performs processing for transmitting a corresponding message to each client such as the client 10 that is a request source (see step S305).

  As described above, the ticket reservation request from the client is stored in a separate request queue for each event, and the reservation status corresponding to each request stored in each request queue can be confirmed. The process of accepting reservations and the process of creating a common message corresponding to each request are configured as a batch process, reducing the processing load on the server that executes the process according to the request for the ticket reservation request be able to.

  In addition, as described above, a category having a possibility of being processed in a batch by common processing is set in advance, and a plurality of requests belonging to the category are collectively stored in a request queue provided corresponding to the category. Since it is configured, part or all of the processing for responding to multiple requests stored in one request queue can be processed simultaneously in one common processing, and the processing load on the server is reduced. It is possible to reduce this, and it is possible to improve the response capability at the time of temporary high load while maintaining the same server capability as during normal operation. Therefore, it is possible to improve the degree of expectation that responses to all requests are completed even during a temporary high load.

  In the second embodiment described above, after all the processes corresponding to each request are completed, the servers 31, 32, and 33 send the processing results (reservation results to the clients such as the client 10). In the case where, for example, the response content for a request stored in a specific request queue is uniquely determined, the request is sent via the queuing device 20. It is also possible to adopt a configuration for replying to the client.

  In this case, when the queuing device 20 receives a common message indicating that the reservation cannot be made from the servers 31, 32, and 33, the queuing device 20 recognizes that the response content to the request is uniquely determined, and the reservation reception of the corresponding event ends. In correspondence with the request queue (request queues 22a, 22b, 22c) or the classification of the information indicating that the reservation has been made and the common message indicating that the reservation cannot be made, in a storage device (an example of a response information storage unit) in the device Keep it. Then, the queuing device 20 thereafter checks the storage state of the request queue (request queues 22a, 22b, 22c) periodically, for example, and stores a request indicating a reservation request for an event for which reservation acceptance has ended. If the request is stored in the request queue (request queue 22a, 22b, 22c), a request message stored in the corresponding request queue is sent to the request source client and a common message indicating that the request cannot be reserved is transmitted. Execute the process to delete.

  If configured as described above, when there is a request for seat reservation with respect to an event in which the seat is full due to the reservation and the reservation reception has been completed, it is possible to respond within the queuing device 20. Therefore, even if a request for reservation is sent for an event for which reservation acceptance has ended, the server transmits a process for confirming the reservation status, a process for creating a common message, and a common message. This eliminates the need to newly perform this process, and can significantly reduce the processing load on the server.

  In the second embodiment described above, all of the plurality of requests stored in the request queues 22a, 22b, and 22c corresponding to the servers 31, 32, and 33 are collectively processed, and all the processes corresponding to the requests are performed. After the processing is completed, the servers 31, 32, and 33 are configured to perform processing for returning the processing result (message indicating the reservation result) to each client such as the client 10 that is the request source. 33 is configured to process the requests stored in the corresponding request queues 22a, 22b, and 22c for each predetermined number (for example, every one or two), for example, a response to a request stored in a specific request queue When the content reaches the specified content, for other requests stored in that specific request queue Answers may be used as the configuration of queuing device 20 is uniquely performed.

  In this case, the servers 31, 32, and 33 read out requests for each predetermined number of requests stored in the corresponding request queues 22a, 22b, and 22c, and part of the predetermined number of requests. Alternatively, a common message for all is generated, and response information including the generated common message is transmitted to the queuing device 20. Specifically, for example, the server 31 reads one request stored first among a plurality of requests stored in the request queue 22a, and determines whether or not to receive a reservation according to the request. Then, response information based on the determination result is transmitted to the queuing device 20. When reply information for one request is transmitted, the next request is read and processed in the same manner. Such processing is executed until there are no more requests in the request queue 22a.

  Further, each time the response information is received from the servers 31, 32, and 33, the queuing device 20 transmits response information including a message indicating the reservation result to the requesting client. However, when the message indicated by the received response information is a message indicating that the message cannot be reserved, the queuing device 20 requests the request source of other requests stored in the corresponding request queue in addition to the request source client. For each client, it is assumed that it is a request received at the same time, and response information including a message indicating that reservation cannot be made is transmitted, and processing for deleting the request stored in the corresponding request queue is executed. . The queuing apparatus 20 stores a newly received request in the request queue after transmitting a message indicating that a common message cannot be reserved to the request source of each request stored in the corresponding request queue. To return to a normal state in which the corresponding server executes processing for generating response information.

  In the above example, the classification is a plurality of requests that can be regarded as requests received at the same time, and the response information generated by the servers 31, 32, and 33 is used in common. What is necessary is just to classify and set beforehand for every request content so that a plurality of requests that may be able to be assigned belong to the same classification. Specifically, for example, in the case of the ticket reservation system as described above, it may be set separately for each same event.

  If configured as described above, when the reservation is completed and the reservation reception ends, and there is another request for a seat reservation for the event for which the reservation reception has ended, it is common that the reservation cannot be made within the queuing device 20 A message can be sent. Therefore, when the reservation acceptance is completed, the server processes the confirmation of the reservation status and the process for creating a common message even if there is another request for reservation for the event for which the reservation acceptance is finished. It is possible to eliminate the need to newly perform a process for transmitting the common message. Therefore, the processing load on the server can be greatly reduced.

  In addition, as described above, since the queuing device 20 is configured to return to the normal state after transmitting the common message, it is stored in the request queue when the seat is already full and no cancellation has occurred. For example, it is only necessary to execute processing at the server for a plurality of requests, and for the other requests, a common message indicating that the queuing device 20 cannot make a reservation can be transmitted. The load can be greatly reduced. In addition, when the seat is full but a cancellation occurs, it is possible to receive a reservation for a request read to the server thereafter until the seat becomes full. That is, when cancellation occurs, a new reservation can be accepted.

  In the second embodiment described above, the request queues 22a, 22b, and 22c are classified for each event that accepts ticket reservation. However, the request queue for each event seat class is used. It is good also as a structure which classifies. With this configuration, the common processing can be executed for each class of event seats, so the server processing load can be reduced even when the event seat reservation process is performed to reserve seats by class. Can be made.

  Hereinafter, as a third example, an example in which the above-described cueing system 100 according to an embodiment of the present invention is applied to a test scoring system 100c for scoring a test will be described. Note that the test scoring system 100c may be a test correction system that determines whether the answer is correct without performing scoring.

  FIG. 10 is a block diagram showing an example of the configuration of the test scoring system 100c. The test scoring system 100c includes a client 10 (not shown in FIG. 10), a queuing device 20, servers 31, 32, and 33, and a test database 42.

  The test database 42 is composed of, for example, a large-capacity storage medium such as a hard disk, and stores various data such as information indicating test questions, information indicating answers to test questions, information indicating test results, and information regarding examinees. Yes.

  In this example, it is assumed that the request queues 22a, 22b, and 22c are classified for each question of the test question. For simplicity, it is assumed that there is only one test question at the same time.

  In this example, each server 31, 32, 33 reads test information from the test database 42 according to the request stored in the corresponding request queue 22a, 22b, 22c, and confirms the answer to the corresponding question. It has a function of determining whether the answer is correct / incorrect, creating scoring information indicating the result, and responding to each client as an examinee.

  Next, request storage processing in this example will be described. In the request storing process, when the class changing unit 23 of the queuing device 20 receives a request from the client 10 (see step S203), it determines the class according to the content of the request (see step S204).

  The request is made by designating one or more questions for which scoring is desired. In this example, the request includes, for example, user information for specifying the address of the client 10, question information indicating each question for which an answer has been created, and answer information indicating an answer to each question. As the question information, for example, an identification number assigned in advance may be used. In this example, it is assumed that each question is a question of a choice formula in order to simplify the explanation.

  Further, the classification of the request is determined by, for example, the classification of the question indicated by the question information in the request. That is, the classification assigned to the question indicated by the question information included in the received request is the request classification.

  When there are a plurality of questions indicated by the question information in the request, the sorter 21 divides the request by question, and divides the classification assigned to the question indicated by the question information included in the divided request. Classification.

  Specifically, as shown in FIG. 11 (A), when the request includes three questions 1 to 3, the request is sent to question 1 as shown in FIG. 11 (B). And the request A including the answer to the question 1, the request B including the answer to the question 2 and the question 2, and the request C including the answer to the question 3 and the question 3. Each of the divided requests A, B, and C includes user information.

  And when classification | category AC is allocated corresponding to each question 1-3 used as the object of scoring service, request A is the classification | category A currently allocated to question 1 among each divided | segmented request. The request B is classified as the category B assigned to the question 2, and the request C is classified as the category C assigned to the question 3.

  Next, the distributor 21 determines a request queue (for example, the request queue 22a) assigned to the determined classification A as a request queue to which the request A is distributed, and scores the question 1 toward the request queue 22a. Request A including the request is transmitted (see step S204). Similarly, request B including the scoring request for question 2 is transmitted toward request queue 22b, and request C including the scoring request for question 3 is transmitted toward request queue 22c.

  Then, the request queues 22a, 22b, and 22c receive the transmitted request and store the received request (see step S205).

  As described above, each request (including divided requests) of a class assigned in advance is stored in each of the request queues 22a, 22b, and 22c.

  Next, request response processing in this example will be described. Here, a case where the server 31 executes a process according to a request stored in the corresponding request queue 22a will be described as an example.

  In the request handling process, the server 31 reads all requests stored in the request queue 22a (see step S301) and executes a common process for all requests (see step S302). In this example, in the common process, scoring for question 1 indicated by each request stored in the request queue 22a is collectively processed.

  Specifically, the server 31 inquires the test database 41 about the correct answer of the question 1. Next, the server 31 performs grouping for each answer content in each request stored in the request queue 22a. In this example, for example, it is divided into a group that selects the correct answer option and a group that selects the incorrect answer option. Further, the server 31 creates a common message indicating correct answer and a common message indicating incorrect answer.

  Next, the server 31 combines the user information (for example, the user name, user ID, answer / correct answer / question answer to the question Q) indicating the requester information and the identification information of the question 1 into the test database 41. To store.

  Further, the server 31 checks whether or not individual processing is necessary, and executes the individual processing if necessary (see steps S303 and S304). For example, as described above, the individual processing may be executed when a message to be returned is created for each client (such as entering a user name).

  When all the processes corresponding to each request are completed, the server 31 sends a common message (for example, “question” to each client who is a request source as a processing result to the correct person as a processing result. 1 is correct ") and a process of transmitting a common message (for example," Question 1 is incorrect ") to the wrong person (see step S305).

  As described above, scoring request requests from clients are stored in separate request queues according to classifications determined individually for each question, scoring processing corresponding to each request saved in each request queue, and each request Since the process of creating a common message corresponding to is performed in a batch process, it is possible to reduce the processing load on the server corresponding to the scoring request.

  In addition, as described above, when a scoring request request from a client includes answers to multiple questions, the request is divided for each question, and each divided request is individually determined for each question in advance. Since each request is stored in a separate request queue, even if multiple requests are included in the request, they can be processed together for each question, and the request for scoring requests can be handled. Server processing load can be reduced.

  In addition, as described above, a category having a possibility of being processed in a batch by common processing is set in advance, and a plurality of requests belonging to the category are collectively stored in a request queue provided corresponding to the category. Since it is configured, part or all of the processing for responding to multiple requests stored in one request queue can be processed simultaneously in one common processing, and the server processing load Can be mitigated, and the response capability at the time of temporary high load can be improved while maintaining the same server capability as during normal operation. Therefore, it is possible to improve the degree of expectation that responses to all requests are completed even during a temporary high load.

  In addition, when the request for scoring about a plurality of questions is made by one request, each server 31, 32, 33 transmits the scoring result to the queuing device 20, and the queuing device 20 sends each server 31, 32, 33 It is also possible to execute a process for collecting the processing results received from 33 for each client, and to collect the scoring results for a plurality of questions and reply to the client. In this case, the message used for the reply may be, for example, “all questions are correct”, “question 1 is correct, question 2 is incorrect, and question 3 is correct”.

It is a block diagram which shows the example of a structure of the queuing system in one embodiment of this invention. It is a flowchart which shows the example of the request storage process in one embodiment of this invention. It is a flowchart which shows the example of the request response process in one embodiment of this invention. It is a block diagram which shows the other example of a structure of the queuing system in one embodiment of this invention. It is a block diagram which shows the example of a structure of the map information system of the 1st Example in one embodiment of this invention. It is explanatory drawing which shows a regional block. It is explanatory drawing which shows a common map and an individual map. It is a block diagram which shows the example of a structure of the ticket reservation system of the 2nd Example in one embodiment of this invention. It is explanatory drawing which shows the example of the reservation status information of a ticket. It is a block diagram which shows the example of a structure of the test scoring system of the 3rd Example in one embodiment of this invention. It is explanatory drawing which shows the example of the division | segmentation state of a request.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Client computer 20 Queuing apparatus 21 Sorter 22a, 22b, 22c Request queue 23 Classification change part 31, 32, 33 Server 40 Map information database 41 Ticket database 42 Test database 100 Queuing system 100a Map information system 100b Ticket reservation system 100c Exam scoring system

Claims (19)

A queuing system for queuing a request received via a communication network and processing a request in the queue in order to respond to the request,
A request distribution unit that identifies a classification of the request from a plurality of classifications according to the content of the received request, and distributes the requests according to the identified classification;
One of the plurality of classifications is assigned, and among the requests distributed by the request distribution means, a plurality of request queues each storing requests of the assigned classification,
Reply processing execution means for executing reply processing for generating reply information for replying to a plurality of requests stored in the request queue for each request queue;
The plurality of classifications are set in advance so as to be classified according to the contents of the request so that a plurality of requests that are executed by a common process in a part or all of the response processes belong to the same classification.
The response processing execution means reads out a plurality of requests stored in the request queue, and performs part or all of the response processing for performing a response to the plurality of requests collectively by the common processing. A characteristic queuing system. The response processing execution means determines whether it is necessary to execute individual processing for each request in the response processing, and if necessary, executes the individual processing and responds to each request. The queuing system according to claim 1, wherein reply information for each is generated. In accordance with the establishment of a predetermined classification change condition, a classification setting change means for changing the setting contents of a plurality of classifications is provided.
The classification setting changing unit executes a process of changing the contents of the classification and changes the setting contents by executing a process of assigning any of the request queues to each of the plurality of changed classifications. The queuing system according to claim 2. The classification change condition is met when the number of requests received per unit time exceeds the specified number.
The queuing system according to claim 3, wherein the classification setting change unit executes a process of reducing the number of classifications and expanding a range of requests belonging to each classification when the classification change condition is satisfied. The classification change condition is satisfied when a predetermined change instruction is given,
The queuing system according to claim 3 or 4, wherein the classification setting change means changes the setting contents in accordance with the change instruction. A plurality of response processing execution means,
The plurality of response processing execution means are each assigned to one of the request queues, execute response processing to reply to the plurality of requests stored in the assigned request queue,
6. The queuing system according to claim 3, wherein the classification changing unit changes the allocation of the request queue in each of the plurality of response processing execution units when a predetermined classification changing condition is satisfied. . The request includes location information for requesting a surrounding map,
Each of the plurality of classifications is divided and set in advance for each position range indicated by the position information,
The response process execution means searches a map database in which map information is stored, and executes a common process for generating a common map including each neighboring map according to a plurality of requests stored in the request queue. The queuing system according to any one of claims 1 to 6. The request includes reservation request information for requesting a ticket reservation,
Each of the plurality of categories is set in advance by dividing each event at least for accepting a ticket reservation,
The response processing execution means searches the ticket database storing the ticket reservation status information, and determines whether or not to accept a reservation for a part or all of the request according to a plurality of requests stored in the request queue. The queuing system according to any one of claims 1 to 6, wherein determination is performed, and common processing is performed to generate a common reservation result message for some or all of the requests according to the determination result. The request includes correction request information for requesting correction of the answer,
Each of the multiple categories is divided into test questions and set in advance.
The response processing execution means searches the ticket database in which the answer information of the test questions is stored, and determines whether or not to correct some or all of the requests according to a plurality of requests stored in the request queue. The queuing system according to any one of claims 1 to 6, wherein determination is performed, and common processing is performed to generate a common correction result message for a part or all of the request according to the determination result. A queuing device that queues requests received via a communication network and distributes and stores the requests in the queue for each predetermined classification in order to reply to the request,
A request distribution unit that identifies a classification of the request from a plurality of classifications according to the content of the received request, and distributes the requests according to the identified classification;
One of the plurality of classifications is assigned, and among the requests distributed by the request distribution means, a plurality of request queues each storing requests of the assigned classifications,
Each of the plurality of request queues performs a response process for reading out the requests stored in the request queue via a communication network and generating response information for replying to the plurality of requests stored in the request queue. Stores the request of the assigned classification in order to cause the response processing execution device to execute to generate the response information for each request queue and execute a part or all of the response processing collectively by common processing. And
The plurality of classifications used by the request distribution means for distribution are the contents of the requests so that a plurality of requests that are partly or wholly executed by the common processing belong to the same classification. A queuing device characterized in that it is classified and set in advance. In accordance with the establishment of a predetermined classification change condition, a classification setting change means for changing the setting contents of a plurality of classifications is provided.
The classification setting changing unit executes a process of changing the contents of the classification and changes the setting contents by executing a process of assigning one of the request queues to each of the plurality of changed classifications. Queuing device. The classification change condition is met when the number of requests received per unit time exceeds the specified number.
The queuing device according to claim 11, wherein the classification setting changing unit executes a process of reducing the number of classifications and expanding a range of requests belonging to each classification when the classification changing condition is satisfied. The classification change condition is satisfied when a predetermined change instruction is given,
The queuing device according to claim 11 or 12, wherein the classification setting changing means changes the setting contents in accordance with the change instruction. Response information receiving means for receiving response information generated for each request queue by the response processing execution device via a communication network;
A response information storage means for storing the received response information and information indicating that when the response content is uniquely determined by the response information in association with the request queue;
When the response content for the request stored in the predetermined request queue is uniquely determined, the response processing execution device does not execute the response processing and is stored in the predetermined request queue. A response information transmission unit that transmits response information stored in the response information storage unit via a communication network toward a request source of each request. The queuing device described. A queuing device that queues requests received via a communication network and distributes and stores the requests in the queue for each predetermined classification in order to reply to the request,
A request distribution unit that identifies a classification of the request from a plurality of classifications according to the content of the received request, and distributes the requests according to the identified classification;
One of the plurality of classifications is assigned, and among the requests distributed by the request distribution means, a plurality of request queues each storing requests of the assigned classification,
From a response process execution device that executes a response process for reading a predetermined number of requests stored in the request queue via a communication network and generating response information for replying to the predetermined number of read requests Response information receiving means for receiving the response information generated for each predetermined number of requests via a communication network;
Response information transmitting means for transmitting the received response information to the request source of the corresponding request via a communication network,
The plurality of classifications used by the request distribution means for distribution are such that the plurality of requests that can be used in common by the response information generated by the response processing execution device belong to the same classification. It is divided and set in advance according to the contents of
When the response content indicated by the received response information is a predetermined content, the response information transmission means is the same as the received response information for each request source stored in the corresponding request queue. A queuing device that transmits response information and executes a process of deleting a request stored in the request queue. A queue processing method for queuing a request received via a communication network and processing a request in the queue in order to respond to the request,
Identifying a classification of the request from a plurality of classifications according to the content of the received request;
Distributing requests according to the identified classification,
One of the plurality of classifications is allocated, and among the requests distributed by the request distribution unit, the requests of the allocated classification are respectively stored in a plurality of request queues;
Reading a plurality of requests stored in the request queue;
Including a step of performing a part or all of response processing for generating response information for replying to a plurality of read requests for each request queue in a batch by common processing,
The plurality of classifications are set in advance so as to be classified according to the contents of the requests so that a plurality of requests executed by a common process in a part or all of the response processes belong to the same classification. A characteristic queue processing method. In response to the establishment of a predetermined classification change condition, a process for changing the contents of the classification is executed, and a process for assigning one of the request queues to each of the plurality of changed classifications is executed, whereby a plurality of classifications are executed. The queue processing method according to claim 16, further comprising a step of changing the setting contents. A queue processing program for queuing a request received via a communication network and processing a request in the queue in order to respond to the request,
On the computer,
Identifying a classification of the request from a plurality of classifications according to the content of the received request;
Distributing requests according to the identified classification,
One of the plurality of classifications is allocated, and among the requests distributed by the request distribution unit, the requests of the allocated classification are respectively stored in a plurality of request queues;
Reading a plurality of requests stored in the request queue;
Performing a step of performing a part or all of the response processing for generating the response information for replying to the plurality of read requests for each request queue in a batch by the common processing,
The plurality of classifications are set in advance so as to be classified according to the contents of the request so that a plurality of requests executed by a common process in a part or all of the response processes belong to the same classification. Feature queue processing program. In addition to executing a process for changing the contents of the classification and a process for assigning one of the request queues to each of the plurality of changed classifications according to the fact that a predetermined classification changing condition is satisfied on the computer. The queue processing program according to claim 18, wherein a step of changing setting contents of a plurality of classifications is executed.

Images