JP2015022554A - Data processing device, data processing program, database system, communication control device, and network system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable shortening of a delay time due to interruption, when data supply from a database server stops and data acquisition is resumed when a data processing device sequentially obtains search processing results from the database server.SOLUTION: There is provided a data processing device. The data processing device includes: means for making a request to a database server for execution of search processing of a database and transmission of the predetermined number of records as one transmission unit data in the order based on a main key; means for processing supplied data; means for holding data of a main key item as final reception position information regarding a final order record of the supplied transmission unit data; and means for requesting execution of search processing for only records not supplied by the database server among the search results and execution of data transmission processing on the basis of the final reception position information after interruption of data supply.

Description

  The present invention relates to a data processing device, a data processing program, a database system, a communication control device, and a network system. For example, a database client that acquires a large amount of search data from a database server (RDBMS (Relational DataBase Management System database client)) It can apply to a communication control apparatus provided with.

  In a conventional database system, when the response result to a transaction (data search request) issued from a database client is a large amount of data, generally, the search result is subdivided into an arbitrary number of records and sent from the database server. In many cases, fetch processing for acquiring data is employed.

  In the fetch function in the conventional database client (the fetch function implemented in the RDBMS), a cursor is opened from the database server for a predetermined number of search results (units for fetch processing), and the cursor opened record is fetched. A large amount of data is acquired sequentially by repeating the (data acquisition) operation.

  In a conventional database client, a buffer (memory) for acquiring a large amount of data can be mounted smaller than the size corresponding to the total number of records of a large amount of search results by this fetch function.

  Conventionally, as a database system (database client) corresponding to the fetch function, for example, there is a technique described in Patent Document 1.

JP2009-245283A

  In the conventional database client, when the cursor open and fetch processing are repeated by the fetch function, a communication session is maintained with the database server, and the cursor position is maintained and managed in the session. Therefore, in the conventional database client, when data acquisition is started by the fetch function, the communication session needs to be maintained until all the retrieval result records are fetched.

  In a conventional database client, if the communication session with the database server is disconnected due to some failure (for example, network failure or database server failure) during data acquisition by the fetch function, the cursor information is discarded. The search result acquisition ends in the middle. This cursor records fetch information related to acquisition target data, that is, which data has been acquired. Therefore, in a conventional database client, if the session is disconnected and the cursor is closed, subsequent data acquisition cannot be performed.

  In such a case, the conventional database client will perform the search again after recovery from the failure and acquire the result again. However, if the number of records in the search result increases, the number of fetches and the acquisition of all data will increase accordingly. Therefore, the time required for the failure also increases, so the risk at the time of the occurrence of the failure is great, and there is a concern that the system may be affected by the fact that the data search process is not completed.

  For example, in a SIP (Session Initiation Protocol) server (for example, a call control device that performs call control of SIP communication) built on a network of a communication carrier, customer data (subscribers) required by the communication control device Data) from a database server. Usually, since there is a large amount of customer data managed by the communication control device of the communication carrier, data acquisition using the fetch function of the database client may be performed in consideration of the processing capability of the communication control device. However, in this communication control device, if there is a failure during acquisition of customer data by the fetch function, there is a possibility that the state where the service cannot be started continues for a long period of time.

  Therefore, when a data processing device (for example, a database client) sequentially acquires search processing results from the database server (for example, when acquiring data by fetch processing), data supply from the database server is stopped and data acquisition is resumed. In this case, a data processing device, a data processing program, a database system, a communication control device, and a network system that reduce the delay time due to interruption are desired.

  The data processing apparatus according to the first aspect of the present invention is (1) executing a search process from a database in which data of a plurality of records are stored, and a primary key comprising one or a plurality of items for a search result by the search process First search requesting means for requesting a database server to perform data transmission processing for transmitting a predetermined number of records as one transmission unit data in the order based on the above, and (2) each transmission unit data supplied from the database server. Data processing means for processing data included in the record, and (3) at least the data of the item of the primary key as the final reception position information for the record in the last order constituting the transmission unit data supplied from the database server A final reception position information holding means to hold, and (4) based on the search processing requested by the first search request means. After the data supply from the database server is interrupted, based on the final reception position information held by the final reception position information holding means, out of the search results of the search processing requested by the first search request means And a second search requesting unit that searches only the record that has not been supplied from the database server and requests the data transmission process for the search result.

  The data processing program according to the second aspect of the present invention comprises (1) a search process from a database in which data of a plurality of records is stored, and a search result by the search process is configured by one or a plurality of items. First search request means for requesting a database server to perform data transmission processing for transmitting a predetermined number of records as one transmission unit data in an order based on the primary key, and (2) each of the data supplied from the database server Data processing means for performing data processing of records included in the transmission unit data; and (3) finally receiving at least the data of the item of the primary key for the records in the last order constituting the transmission unit data supplied from the database server. A final received position information holding means for holding as position information; and (4) the first search request means. After the data supply from the database server based on the requested search process is interrupted, the search process requested by the first search request means based on the last received position information held by the last received position information holding means Of these search results, only a record that has not been supplied from the database server is searched, and the search result is made to function as a second search request means for requesting the data transmission process.

  In a third aspect of the present invention, in a database system including one or a plurality of database servers and a data processing apparatus that requests the database server to perform a search process, the data processing apparatus of the first aspect of the present invention is applied as the data processing apparatus. A database system characterized by that.

  A communication control device according to a fourth aspect of the present invention includes a data processing device that acquires data related to communication control from a database server, and a communication control process that performs communication control processing between the communication devices using the data acquired by the data processing device. A communication control apparatus comprising: a data processing apparatus according to the first aspect of the present invention as the data processing apparatus.

  In a fifth aspect of the present invention, in a network system including a plurality of communication devices and a communication control device that performs communication control processing between the communication devices, the communication control device of the fourth aspect of the present invention is applied as the communication control device. A network system characterized by this.

  According to the present invention, when the data processing apparatus sequentially acquires the search processing results from the database server, when the data supply from the database server is stopped and the data acquisition is restarted, the delay time due to the interruption can be reduced. it can.

1 is a block diagram showing an overall configuration of a network system according to an embodiment. It is explanatory drawing shown about the structural example of the customer database which the database server which concerns on embodiment has. It is explanatory drawing shown about the SQL sentence (when there is one primary key) of the search request which the database client which concerns on embodiment transmits to a database server. It is the sequence figure shown about the data acquisition sequence (normal operation | movement) between the database client which concerns on embodiment, and a database server. It is explanatory drawing shown about the search result (when there is one primary key) of the customer database which the database server which concerns on embodiment acquires. It is the sequence figure shown about the data acquisition sequence (when interruption generate | occur | produces the 1) between the database client which concerns on embodiment. It is the sequence figure shown about the data acquisition sequence (when interruption generate | occur | produces the 2) between the database client which concerns on embodiment. It is explanatory drawing shown about the SQL sentence (when there is one primary key) produced | generated when the database client which concerns on embodiment restarts a data acquisition sequence. It is explanatory drawing shown about the search result (when there is one primary key) of the customer database which the database server which concerns on embodiment acquires when a data acquisition sequence restarts. It is explanatory drawing shown about the SQL sentence (when there are multiple primary keys) of the search request which the database client which concerns on embodiment transmits to a database server. It is explanatory drawing shown about the search result (when there are multiple primary keys) of the customer database which the database server concerning an embodiment acquires. It is explanatory drawing shown about the SQL sentence (when there are two or more primary keys) produced | generated when the database client which concerns on embodiment restarts a data acquisition sequence. It is explanatory drawing shown about the production | generation logic of the SQL sentence (when there are two or more primary keys) which the database client which concerns on embodiment produces | generates when restarting a data acquisition sequence. It is explanatory drawing shown about the SQL sentence which the database client of the modification which concerns on embodiment produces | generates.

(A) Main Embodiment Hereinafter, an embodiment of a data processing device, a data processing program, a database system, a communication control device, and a network system according to the present invention will be described in detail with reference to the drawings. In the following, an example in which the data processing apparatus of the present invention is applied to a database client will be described.

(A-1) Configuration of Embodiment FIG. 1 is a block diagram showing the overall configuration of the network system of this embodiment. Further, FIG. 1 also shows a functional configuration of the communication control apparatus 10 of this embodiment.

  The network system 1 includes a communication control device 10, two database servers 20 (20-1 and 20-2), and a plurality of communication devices 30 connected under the communication control device 10. Note that the number of devices constituting the network system 1 is not limited. In addition, the communication control device 10 and the database servers 20-1 and 20-2 are connected to the network N1, respectively, and are configured to be able to communicate with each other. Each is connected to the network N3 and can communicate with each other.

  The communication control device 10 is a device that performs communication control of communication between the communication devices 30 (for example, call control by SIP message exchange).

  Each communication device 30 communicates with another communication device 30 based on the communication control of the communication control device 10. As each communication device 30, for example, a PC or a smart phone installed with software corresponding to a voice call or a video call based on SIP message exchange can be applied.

  The database servers 20-1 and 20-2 are servers that provide one database service made redundant by two units.

  Next, the detailed configuration of the database servers 20-1 and 20-2 will be described.

  The database servers 20-1 and 20-2 have customer data necessary for the communication control device 10 to perform communication control processing (information relating to each communication device 30 itself or a customer (subscriber) using each communication device 30, etc. Data) is managed as a database, and the managed database is provided to the communication control device 10. The database servers 20-1 and 20-2 are operating in a redundant configuration (hot standby system) in which either one operates as an active system (active state) and the other operates as a standby system (standby state). And In the following description, it is assumed that the database server 20-1 normally operates as an active system and the database server 20-2 operates as a standby system. The specific method of the redundant configuration of the database servers 20-1 and 20-2 is not limited to the hot standby described above, and various redundant configurations such as a cold standby method and a clustering method can be applied.

  Each database server 20 has the same configuration, and functionally includes a server process 21 that performs data processing of a database service and a file system 22 that functions as data storage of the database.

  The server process 21 can be constructed using, for example, various RDBMS server platforms. The file system 22 can use various data storage (for example, data storage using a hard disk drive or the like). In this embodiment, the server process 21 will be described as being constructed by an RDBMS that can accept data processing (data search) based on a command (program) using SQL (Structured Query Language) from a database client.

  Further, it is assumed that the file system 22 of the database server 20-1 and the file system 22 of the database server 20-2 are connected via the network N2, and data synchronization is performed in real time. Since the data synchronization configuration between the database servers 20 can be applied to the same configuration as various data storages, detailed description thereof is omitted. As described above, the file system 22 stores a database of customer data to be provided to the communication control apparatus 10.

  In this embodiment, data of a structure that can be represented by a table as shown in FIG. 2 is stored as a customer data database (hereinafter also referred to as “customer database”) stored in the file system 22 of the database server 20. Explain that it is. The specific structure of the customer database stored in the file system 22 of the database server 20 is not limited.

  As shown in FIG. 2, in the customer database, data represented by one record (one line of data) is stored for one customer. As shown in FIG. 2, in this embodiment, each record includes at least customer registration number, name, sex, address, and telephone number data relating to the customer. The items to be configured are not limited. For example, items constituting each record may include a user ID, password, address information (identification information of the corresponding communication device 30) used for various systems, and the like.

  Next, a detailed configuration of the communication control apparatus 10 will be described.

  As shown in FIG. 1, the communication control device 10 functionally includes a database client 11, a customer data storage unit 12, and a communication processing unit 13. As the communication control apparatus 10, for example, software (a program including the data processing program of the present invention) corresponding to the database client 11 and the communication processing unit 13 is installed in an existing computer such as a PC or a workstation. It may be realized by.

  The database system of this embodiment will be described as being configured by the database client 11 of the communication control apparatus 10 and the database servers 20-1 and 20-2. In this embodiment, the database client 11 is described as being mounted on the communication control device 10, but the database client 11 may be constructed as a single device.

  The customer data storage unit 12 is a storage unit for storing each record (customer data) acquired by the database client 11 requesting the database servers 20-1 and 20-2 to perform a search process.

  The communication processing unit 13 functions as a communication control unit that performs communication control between the communication devices 30 using an SIP message or the like. It is assumed that the communication processing unit 13 uses customer data stored in the customer data storage unit 12 during communication control. The protocol used by the communication processing unit 13 for communication control and the items of customer data scattered by the communication processing unit 13 during communication control are not limited. Since the configuration of the communication processing unit 13 itself can be the same as that of various communication control devices (for example, a SIP server), detailed description thereof is omitted.

  The database client 11 acquires part or all of the customer database from the database server 20 (20-1, 20-2) by the search request process, and registers / updates customer data stored in the customer data storage unit 12. Process. When the communication control device 10 is realized by software, the program corresponding to the database client 11 corresponds to the data processing program of the present invention.

  The database client 11 may be realized, for example, by operating a user program on a database application (middleware) provided by various RDBMS platforms. In this embodiment, the database client 11 is functionally described as being configured using three components (program, module) of the database processing unit 111, the pointer holding unit 112, and the data acquisition continuation processing 113. However, the specific module configuration for realizing each component is not limited. In this embodiment, in order to simplify the description, the database client 11 is described as being configured using the above-described three components, but the specific module structure is not limited. For example, the database client 11 may be configured by incorporating all the functions of the above three components in one module.

  The database processing unit 111 mainly requests the database server 20 to perform a search process on the customer database, and processes the search result (in this embodiment, stores the acquired search result as customer data in the customer data storage unit 12) ) Specifically, the database processing unit 111 requests the database server 20 (server process 21) to perform a search process using the fetch function (search process for the customer database), and obtains the search result (data of the corresponding record). Receive supply. Here, it is assumed that 100 is set as the number of records to be acquired (hereinafter referred to as “fetch number”) in one fetch process in the database processing unit 111. In other words, the database processing unit 111 requests the database server 20 (server process 21) to sequentially transmit a record of the number of fetches in the search result as one transmission unit. Note that the number of fetches used when the database processing unit 111 acquires data may be dynamically changed. However, the database processing unit 111 of this embodiment is described as using only the initially set number of fetches. To do.

  In this embodiment, the database processing unit 111 is described as performing a search processing request to the database server 20 (server process 21) using an SQL statement. However, a specific database language or the like used for the search processing request is described. The contents of the sequence are not limited. For example, the database client 11 uses the SQL statement (SQL statement using the SELECT command) “SELECT customer registration number, name, gender, age, address, phone number FROM customer data ORDER BY customer registration number” shown in FIG. It can be used to request customer data from the database server 20 (server process 21).

  The SQL statement shown in FIG. 3 includes “customer registration number” as a primary key for data items of “customer registration number, name, gender, age, address, telephone number” in the “customer database” table of the database server 20. It shows that it is acquired in the sorted order as (primary key). Hereinafter, items to be extracted in the SQL sentence (items following SELECT, for example, “customer registration number, name, gender, age, address, phone number” in FIG. 3) are referred to as “extraction items”. And

  Therefore, when the SQL statement in FIG. 3 is issued from the database client 11, the database server 20 (server process 21) extracts the extracted items (“customer registration number, name, gender, age, address” from the customer data. , Phone number ") is extracted (for example, expanded in a memory), and the search result is extracted by the number of fetches (100 records) by the database client 11. Will be sent to. Further, while the communication session is maintained between the database client 11 and the database server 20 (server process 21), the fetched position (cursor) is automatically updated by a function provided by the middleware. . Hereinafter, a series of processing (sequence) in which the database client 11 connects to the database server 20 (server process 21) using a communication session and sequentially acquires data by fetch processing is referred to as “data acquisition sequence”. Shall. Since various RDBMS processes can be applied to the normal data acquisition sequence between the database client 11 and the database server 20 (server process 21), detailed description thereof is omitted.

  The pointer holding unit 112 holds the last fetched position (cursor position) as pointer information each time the database processing unit 111 fetches data from the database server 20 (acquires data of records corresponding to the number of fetches). . Specifically, the pointer holding unit 112, when the database processing unit 111 acquires the data of the number of records for the number of fetches in one fetch process, the last record acquired in the fetch process (that is, the acquired fetch Among several minutes of records, data (at least the data of the primary key) of the record that is the last order when sorted by the primary key is held as “pointer information”.

Then, the database processing unit 111 interrupts the data acquisition sequence due to a communication failure or the like (communication session is disconnected) while acquiring customer data from the database server 20-1 (server process 21) by the data acquisition sequence using the fetch function. ), The acquisition of customer data can be resumed from the database server 20-2 (server process 21) that will newly operate as the active system. At this time, the database processing unit 111 creates an SQL statement for restarting the data acquisition sequence from the next record at the cursor position fetched last time based on the pointer information held in the pointer holding unit 112, and The server 20-2 (server process 21) is requested to resume the data acquisition sequence. In other words, at this time, when the data acquisition sequence is interrupted, the database processing unit 111 generates an SQL statement for requesting data transmission of a record that cannot be acquired due to the interruption in the search result, and the database server 20- 2 (Server Process 21) (A-2) Operation of Embodiment Next, the operation of the network system 1 of this embodiment having the above configuration will be described.

(A-2-1) Normal operation of data acquisition sequence First, the operation when the communication control device 10 (database client 11) completes the data acquisition sequence with the database server 20-1 normally (when no failure occurs) Will be described with reference to the sequence diagram of FIG. In the initial stage, it is assumed that the contents of the customer data storage unit 12 of the communication control device 10 are in an initialized state (no record is registered).

  First, the database client 11 of the communication control apparatus 10 requests the database server 20-1 (server process 21) to generate a communication session (communication session provided by middleware), and the database server 20-1 (server It is assumed that a communication session with the process 21) has been established (S101, S102).

  Next, it is assumed that the database client 11 makes a search request (request for database processing using a SELECT statement) to the database server 20-1 (server process 21) using the SQL statement shown in FIG. (S103). At this time, it is assumed that 100 is set as the number of fetches in the communication session between the database client 11 and the database server 20-1 (server process 21). Further, as shown in FIG. 3 described above, the primary key of the customer data search request is a registered customer number.

  Then, the database server 20-1 (server process 21) that has received the SQL statement performs a search process (SQL statement processing using the SELECT statement) based on the received SQL statement for the customer database of the file system 22. The search result is held (S104).

  At this time, the search results held in the database server 20-1 (server process 21) are represented in a table format as shown in FIG. In the table shown in FIG. 5, the records (rows) having the data of each extraction item are sorted in the order based on the primary key (customer registration number). Further, the table shown in FIG. 5 has data of 10,000 records with customer registration numbers 1 to 10,000. The database server 20-1 (server process 21) performs processing for transmitting (fetching) the search result table to the database client 11 by 100 records from the top.

  Then, the database server 20-1 (server process 21) acquires the first 100 records (records with the first to 100th sort order based on the primary key) from the search result, and transmits them to the database client 11 (S105). .

  Then, when the database client 11 fetches (acquires) the data of the first 100 records, the data of the 100 records is processed (added to the customer data storage unit 12) (S106).

  Further, when the database client 11 fetches (acquires) the data of the first 100 records, it supplies the pointer holding unit 112 with the primary key value of the last record (the last record in the primary key sort order) among the fetched records. (S107). At this time, since the database client 11 fetches the records with the registration numbers 1 to 100, the registration number 100 corresponding to the last is supplied to the pointer holding unit 112 and held therein.

  Next, the database client 11 transmits the next fetch request (second fetch request) to the database server 20-1 (server process 21) (S108).

  Upon receiving the next fetch request (second fetch request), the database server 20-1 obtains the next 100 records (records with a 101 to 200th sort order based on the primary key) from the search result, and The data is transmitted to the client 11 (S109).

  Then, when the database client 11 fetches (acquires) the data of the next 100 records (the 101st to 200th records based on the primary key), the data of the 100 records is processed (added to the customer data storage unit 12). Further, the primary key value (registered customer number = 200) of the last record among the fetched records is supplied to the pointer holding unit 112 and held (S111).

  Thereafter, the data acquisition sequence is continued between the database client 11 and the database server 20-1 (the fetch processing in steps S108 to S111 described above is repeatedly executed), and the database client 11 receives the database server 20-1 (server process 21). ) To the 100th fetch request (fetch request) (S201).

  Upon receiving the 100th fetch request, the database server 20-1 acquires the last 100 records (the records with the sort order of 9901 to 10000 based on the primary key) from the search result, and transmits them to the database client 11 ( S202). At this time, the database server 20-1 also notifies the database client 11 that record transmission of all search results is completed by the current fetch.

  When the data fetching (acquisition) of the last 100 records is completed, the database client 11 processes the data of the 100 records (additionally records in the customer data storage unit 12) (S203), and further, the pointer holding unit 112 Is initialized (discarded) (S204).

  Further, when the data fetch (acquisition) of the last 100 records is completed, the database client 11 transmits a session release (session disconnection request) to the database server 20-1 (S205).

  Upon receiving the session release (session disconnection request), the database server 20-1 executes the session release and returns a response indicating that the session has been released to the database client 11 (S206).

  In the case of a normal system, the data acquisition sequence in the database client 11 is performed according to the above sequence.

(A-2-2) Abnormal operation of data acquisition sequence Next, while the database client 11 is executing the data acquisition sequence with the database server 20-1, a failure occurs in the database server 20-1, and the data The operation when the acquisition sequence is interrupted will be described with reference to the sequence diagrams of FIGS.

  In the sequence diagrams of FIGS. 6 and 7, the same step numbers (symbols) are assigned to the same processing steps as those in the sequence diagram of FIG.

  First, in the same manner as the normal operation described above, the database client 11 starts a data acquisition sequence with the database server 20-1 (server process 21) and starts fetch processing (steps S101 to S111 described above). It is assumed that the fetch process is continuously repeated.

  After that, the database client 11 requests the database server 20-1 (server process 21) to execute the 50th fetch process (S301), and acquires data of records (100 records) corresponding to the number of fetches (S302). ), Additionally registered in the customer data storage unit 12 (S303). At this time, the database client 11 supplies the pointer holding unit 112 with the primary key value (registered customer number = 5000) of the last record among the fetched records (S304).

  Then, after the 50th fetch process by the database client 11 is completed and before the next fetch process is completed, it is assumed that a failure has occurred in the database server 20-1 and the operation is stopped (S401).

  Thereafter, it is assumed that the database server 20-2 detects a failure of the database server 20-1 and switches from the standby system to operate as the active system (active system) (S402).

  Thereafter, the database client 11 recognizes that a failure has occurred in the database server 20-1, the data acquisition sequence has been interrupted, and that the database server 20-2 has been switched to the active system. It is assumed that it is switched to 2 (S403).

  It should be noted that various redundant database system configurations can be applied to the switching of the active system from the database server 20-1 to the database server 20-2 and the server switching configuration in the database client 11. Detailed description is omitted. For example, when the communication session between the database client 11 and the database server 20-1 is managed by TCP, for example, the database client 11 cannot transmit and receive packets in the communication session (TCP session) for a certain period of time or longer. In this case, the communication session disconnection can be detected.

  Next, based on the pointer information (customer registration number = 5000) held in the pointer holding unit 112, the database client 11 restarts the data acquisition sequence from the record next to the cursor position fetched last time. Is created (S404).

  Specifically, the database client 11 generates the SQL statement shown in FIG. 8 based on the pointer information (customer registration number = 5000) held in the pointer holding unit 112. In the SQL statement shown in FIG. 8, as a search condition for the SELECT statement, a search for extracting records in which the sort order is after the pointer information (customer registration number = 5000) in the initial SQL statement (the SQL statement in FIG. 3 described above). An SQL statement to which the condition “WHERE customer registration number> 5000” is added is generated.

  Then, the database client 11 of the communication control device 10 issues a communication session generation request to the database server 20-2 (server process 21), and creates a session with the database server 20-2 (server process 21). Establish (S405, S406).

  Next, the database client 11 makes a search request to the database server 20-2 (server process 21) using the SQL statement (FIG. 8 described above) generated in step S404 described above (S407).

  The database server 20-2 (server process 21) that has received the SQL statement performs a search process based on the SQL statement received from the customer database of the file system 22, and holds the search result (S408).

  At this time, an example of a search result held in the database server 20-2 (server process 21) is shown in a table format as shown in FIG. At this time, the search result held in the database server 20-2 (server process 21) is composed of data of 5000 records with customer registration numbers 5001 to 10000 as shown in FIG. The database server 20-2 (server process 21) performs the process of transmitting (fetching) 100 records to the database client 11 for each of the search result tables 50 times.

  Then, the database server 20-2 (server process 21) obtains the first 100 records from the search result (the records with the first to 100th sort order based on the primary key and the records with customer registration numbers 5001 to 5100). The data is sent to the database client 11 (S409).

  When the database client 11 fetches (acquires) the data of 100 records, the data of the 100 records is processed (added to the customer data storage unit 12) (S410), and the last record among the records fetched this time The primary key value (registered customer number = 5100) is supplied as pointer information to the pointer holding unit 112 and held (S411).

  After that, the database client 11 requests the database server 20-2 (server process 21) to perform the second fetch process after the restart (S412), and records for the number of fetches (customer registration numbers are 5111 to 5200). 100 records) is acquired (S413) and additionally registered in the customer data storage unit 12 (S414). At this time, the database client 11 supplies the primary key value (registered customer number = 5200) of the last record among the records fetched this time to the pointer holding unit 112 as pointer information to hold it (S415).

  Thereafter, the data acquisition sequence is continued between the database client 11 and the database server 20-2 (the above-described steps S411 to S415 are repeatedly executed), and the database client 11 is transferred to the database server 20-2 (server process 21). In contrast, it is assumed that the 50th fetch request (fetch request) has been transmitted after the restart (S501).

  Upon receiving the 50th fetch request after the restart, the database server 20-2 obtains the last 100 records (records having a sort order of 9901 to 10000 based on the primary key) from the search result, and transmits them to the database client 11. (S502).

  Then, when the fetching (acquisition) of the data of the last 100 records is completed, the database client 11 processes the data of the 100 records (added recording to the customer data storage unit 12) (S503), and further, the pointer holding unit 112 Is initialized (discarded) (S504).

  Further, when the data fetch (acquisition) of the last 100 records is completed, the database client 11 transmits a session release (session disconnection request) to the database server 20-2 (S505).

  Upon receiving the session release (session disconnection request), the database server 20-2 executes the session release and returns a response indicating that the session has been released to the database client 11 (S506).

(A-2-3) When the primary key is composed of a plurality of items In the above example, the primary key for the communication control device 10 (database client 11) to determine the order in which customer data is acquired is 1 Although only one item (customer registration number) is used, a plurality of items may be applied to the primary key when actually processing the subscriber data.

  Below, the operation example of the network system 1 in case a primary key is comprised by two or more items is demonstrated.

  For example, if there is a record of users who are in the same household in the customer data, the household is considered to have the same address and telephone number (in the case of a fixed telephone). An example SQL sentence for obtaining a search result obtained by sorting each record of customer data for each household is as shown in FIG.

  The SQL sentence shown in FIG. 10 describes “ORDER BY address, telephone number, name” and describes that three primary keys are applied in the order of “address, telephone number, name”. However, here, for the sake of simplicity, it is assumed that there is no record of a user having the same family name and the same name (name character strings are exactly the same) in the same household of customer data.

  Then, when a data search request using the SQL statement shown in FIG. 10 is supplied from the database client 11 to the database server 20, the database server 20 obtains an example of a search result as shown in FIG. .

  According to this example, “ORDER BY address, telephone number, name”, which is the condition of the SQL sentence, is “sorting in ascending order by address, sorting by telephone number, and sorting by name”. Therefore, when the search result as shown in FIG. 11 is obtained, the record of the customer number 00002 and the record of the customer number 0800 can be regarded as the same household because the address and the telephone number match. Further, when the search as shown in FIG. 11 is obtained, the user of the record with the customer registration number 00001 and the user of the record with the customer registration number 00003 can be regarded as the same household.

  When the database client 11 performs the data acquisition sequence (repetition processing of fetch processing) with the SQL statement shown in FIG. 10, every time the fetch number of records is fetched, the primary key of the record acquired last by the fetch Are stored as “pointer information” (that is, data of address, telephone number, and name items).

  Here, it is assumed that the pointer information held by the database client 11 (pointer holding unit 112) has the contents of “address = xxxxxxx”, “phone number = nnnnnnnnnn”, “name = ZZZZ”. Suppose.

  In this case, FIG. 12 shows an SQL sentence generated by the database client 11 based on the pointer information after the data acquisition sequence is resumed.

  When the database client 11 resumes the data acquisition sequence, first, as shown in FIG. 12A, the condition “WHERE address = xxxxxxx AND telephone number = nnnnnnnnnn name> ZZZZ” is added to the search condition. An SQL statement is generated and a data acquisition sequence is executed (the process of fetching from the database server 20 is repeated until there is no search result).

  Next, as shown in FIG. 12B, the database client 11 generates an SQL statement by adding a condition of “WHERE address = XXXXXX AND telephone number> nnnnnnnnnn” to the search condition to obtain data. Run the sequence.

  Finally, as shown in FIG. 12C, the database client 11 generates an SQL statement in which a condition “WHERE address> xxxxxxx” is added to the search condition, and executes the data acquisition sequence.

  In this example, as described above, since the primary key is composed of a plurality of items in the order of “address, telephone number, name”, the database client 11 can execute the above three SQL statements after restarting the data acquisition sequence. By executing, the customer data storage unit 12 can hold the same search result as the search result in the case of normal operation (the search result of the SQL sentence in FIG. 10 described above).

  For example, in the database client 11, an example in which the number of fetches is “1” and a data acquisition sequence is performed using the above-described SQL statement in FIG. 10 will be described. In this case, the search result acquired by the SQL sentence of FIG. 10 is similarly the content of FIG. At this time, it is assumed that a failure occurs after fetching up to the customer registration number “00001” in the search result in the database client 11 and the data acquisition sequence is interrupted. At this time, the pointer holding unit 112 holds information such as “address = 1-1-1 B-cho, A-ku, Tokyo, phone number = 03-1111-1111, name = Taro Sato” as pointer information.

  In this case, since there is a possibility that data of the same household as the customer registration number “00001” exists in the search result, when the database client 11 performs a continuous search, first, “the address is the same” In addition, the telephone number is the same, the character string of the name is sorted, and the order is later than the name of the customer number 00001 ”(search by the SQL sentence in FIG. 12A described above) is performed. There is a need. At this time, if a corresponding record (a record of another user in the same household) exists, the database client 11 first fetches the data of the record in the continuous search process and adds it to the customer data storage unit 12. If not, it is against the specification of “search by household”.

  In general, in database searches, registered character strings are handled as binary codes and sorted in ascending order (or descending order) of their values, so simple numeric comparison is a search method. Is different. A specific method of comparing character strings in the database server 20 is not limited, and methods based on various code systems can be applied.

  When the data acquisition sequence is completed in the search based on the first condition after the restart (search by the SQL statement in FIG. 12A) ("data acquisition of all records" or "no corresponding record") Next, the database client 11 records “the address is the same and the telephone number is in the search order after the telephone number (= 03-1111-1111) of the last fetched data at the time of failure”. Must be searched (search by the SQL sentence in FIG. 12B described above). The search based on the second condition after the restart (search by the SQL sentence in FIG. 12B) may be performed in the same household, for example, when the phone number at the time of customer data registration is a mobile phone number. It is necessary processing to be performed.

  Then, when the data acquisition sequence is completed in the second search after the restart (search by the SQL statement in FIG. 12B), the database client 11 can acquire all the unacquired data of the same household. Therefore, since the next search is a data search after the next household on the original generation list, the search order after “address of pointer information (= 1-1-1 B-cho, A-ku, Tokyo)” ”Records (search by the SQL sentence in FIG. 12C described above)).

  As a result, after the data acquisition sequence is resumed, the database client 11 sequentially executes the search processing (data acquisition sequence) using the three SQL statements described above, and adds the search results to the customer data storage unit 12 (the above-described diagram). 6 and the same processing as in FIG. 7, etc., it is possible to obtain the same result as in the normal operation together with the data acquired before the interruption.

  Based on the above-described specific example, the SQL statement search condition (description after WHERE) generated by the database client 11 when the data acquisition sequence is resumed is shown in a generalized form as shown in FIG.

  FIG. 13A shows the pointer information when the columns (items) of the n primary keys are “C1, C2,..., Cn−2, Cn−1, Cn”, respectively. 13A, the pointer information held by the pointer holding unit 112 for each of the n columns “C1, C2,..., Cn−2, Cn−1, Cn” is “V1, V2,. Vn-2, Vn-1, Vn ". For example, when the primary key is “address, phone number, name” as in the example of FIG. 10 described above, column C1 (V1) is “address”, column C2 (V2) is “phone number”, column C3 ( V3) becomes “name”, and the number of columns n = 3.

  In the generalized example of FIG. 13, the search conditions (descriptions after the WHERE) of n SQL statements generated when the database client 11 resumes the data acquisition sequence are the contents of FIG. 13B, respectively. .

  In FIG. 13B, in the first search condition after the data acquisition sequence is restarted, the values of the pointer information (V1 to Vn-1) and the equal condition “=” are set in the WHERE clause for the columns C1 to Cn-1. Further, the column Cn is assigned to the WHERE clause as a condition of> (greater than) the value Vn of the pointer information.

  In FIG. 13B, in the second search condition after the data acquisition sequence is restarted, the values of the pointer information (V1 to Vn-2) and the equal condition “=” are set for the columns C1 to Cn-2. The column Cn−1 is added to the WHERE clause as a condition “>” (greater than) for the pointer information value Vn−1.

  The database client 11 repeats the generation of the search condition with the above assignment logic.

  As shown in FIG. 13B, in the (n-1) th search condition after resuming the data acquisition sequence, the column C1 is assigned to the WHERE clause as the pointer information value (V1) and the equal condition “=”. The column C2 is assigned to the WHERE clause as a condition “>” (greater than) with respect to the value V2 of the pointer information.

  Furthermore, as shown in FIG. 13B, in the n-th search condition after the data acquisition sequence is resumed, the WHERE clause is set as a condition “>” (greater than) for the value V1 of the pointer information for the column C1. Has been granted.

  Even when the configuration of the primary key changes, the database client 11 (database processing unit 111) generates an SQL statement in which search conditions are set in the order shown in FIG. 13, and executes a data acquisition sequence using each SQL statement. can do.

  In the phrase “ORDER BY” of the normal SELECT command, columns (items) used for sorting are described in descending order of priority. For example, since “address, phone number, name” is described after “ORDER BY” in the example of FIG. 10 described above, “address” is set as the highest priority column (item) in the primary key. ing. This is because the address is the most important key (column) for classifying households in customer data (subscriber data). Therefore, in the example of FIG. 12 described above, the database client 11 searches for a record of a user having the same address (household) in the SQL sentence of the first search condition and the SQL sentence of the second search condition after the data sequence is resumed. After that, a search process is performed for a record whose search order is later than the address (household) in the SQL statement of the third search condition. The same processing will be described using the generalized example of FIG. 13. The database client 11 uses the highest priority column C1 = V1 (in the SQL statement of the first to n−1 search conditions after the data acquisition sequence is restarted. Search for unacquired records (records that have not been fetched yet) with the same pointer information value), and search for records whose column C1 is the search order after V1 in the SQL statement of the nth search condition It is processing.

  Note that the SQL statements shown in FIGS. 12 and 13 are merely specific examples of the above-described processing. If the same processing is possible in accordance with the specifications of the middleware (database application) to be used, the description content is different. May be.

(A-3) Effects of Embodiment According to this embodiment, the following effects can be achieved.

  When the data acquisition sequence is interrupted, the database client 11 generates an SQL statement for requesting data transmission of a record that cannot be acquired due to the interruption in the search result, and sends it to the database server 20-2 (server process 21). Transmitting and restarting the data acquisition sequence. As a result, the database client 11 can obtain a result equivalent to that during normal operation by combining the search results acquired before the interruption of the data acquisition sequence and the search results acquired after the data acquisition sequence is resumed. Therefore, the database client 11 changes the search condition each time the data acquisition sequence is interrupted a plurality of times before data acquisition is completed for all search results (a pointer storage unit in the SQL statement). By simply setting a search condition based on the pointer information held in 112, a continuous data acquisition sequence is possible. That is, in the case where the data acquisition sequence is interrupted in the database client 11, the delay time due to the interruption is the operation of restarting the data acquisition sequence (establishing a communication session with the database server, processing for generating and transmitting a new SQL sentence, etc. ) Only occurs, and can be made shorter than before.

  In particular, since the number of registered records of customer data (subscriber data) used in communication control devices (SIP servers, etc.) of communication carriers is enormous, in the conventional technology, for example, data acquisition is performed for record registration in units of one million items. The delay time when the sequence is performed from the beginning may be very long. In addition, since it is difficult for a communication control device of a communication carrier to start a service when customer data is not available, a delay in the data acquisition sequence may affect the carrier service. Therefore, when the database client (data processing apparatus, data processing program) of the present invention is used for a communication control apparatus of a communication carrier, a greater effect can be achieved.

(B) Other Embodiments The present invention is not limited to the above-described embodiments, and may include modified embodiments as exemplified below.

(B-1) In the above embodiment, in order to simplify the description, an example in which there is no initial search condition (initial WHERE clause) added to the SQL statement used in the data acquisition sequence of the database client 11 has been described. Any search condition may be assigned from the beginning.

  In addition, in the search condition to which this method is applied, a WHERE clause for narrowing down the data search range may be arbitrarily given, and the sort condition item given as the applicable condition is narrowed down by an arbitrarily given WHERE clause column. This is an item for obtaining uniqueness within the specified data range.

  FIG. 14 is an explanatory diagram showing an example of an SQL sentence (an SQL sentence to which a search condition is given from the beginning) used by the database client 11 in a data acquisition sequence.

  FIG. 14A shows an initial SQL statement in the data acquisition sequence. FIG. 14B shows an SQL statement when the data acquisition sequence is restarted. In FIG. 14, the primary key is only “customer registration number”.

  In FIG. 14 (a), as an example, in the WHERE clause, “WHERE gender = male AND age> = 30 AND address as a search condition for searching under the condition“ male, age 30 years or older, living in Tokyo ” "LIKE Tokyo".

  When the database client 11 is interrupted while executing the data acquisition sequence using the SQL statement in FIG. 14A and restarts, as shown in FIG. 14B, “WHERE gender = male An SQL statement to which a search condition “AND age> = 30 AND address LIKE Tokyo AND customer registration number> nnnn” can be used. Note that “nnnn” in the SQL sentence in FIG. 14B is the value of the pointer information held in the pointer holding unit 112 (the value of the primary key “customer registration number”).

(B-2) In the above embodiment, the database client (data processing apparatus, data processing program) of the present invention is applied to the processing of customer data (subscriber data) of the communication control apparatus. The usage of (data processing device, data processing program) is not limited, and can be applied to all systems that "search from a database under specific conditions and obtain the result by multiple fetches" It is.

(B-3) In each of the above embodiments, only the process of additionally recording the data fetched from the database server 20 by the database client 11 in the customer data storage unit 12 is performed, but various processing processes are performed on the fetched data. (For example, processing for converting the display format and contents of data) may be performed. In that case, in the database client 11 of the present invention, the effect of shortening the delay time when interruption occurs in the data acquisition sequence is greater.

(B-4) In the above embodiment, the database server has been described as being redundant. However, the database server may not be redundant. In this case, when the data acquisition sequence is interrupted, the database client resumes the data acquisition sequence with the same database server after restoration of the service of the database server (for example, after restoration of a communication failure or the like).

  DESCRIPTION OF SYMBOLS 1 ... Network system, 10 ... SIP server, 20, 20-1, 20-2 ... Database server, 21 ... Server process, 22 ... File system, 10 ... Communication control apparatus, 11 ... Database client, 111 ... Database processing part, DESCRIPTION OF SYMBOLS 112 ... Pointer holding part, ... Customer memory part, 13 ... Communication processing part, 30 ... Communication apparatus, N1, N2, N3 ... Network.

Claims (6)

A search process is executed from a database in which data of a plurality of records is stored, and a predetermined number of records are transmitted as one transmission unit in an order based on a primary key composed of one or a plurality of items for a search result by the search process. First search requesting means for requesting a database server for data transmission processing to be transmitted as data;
Data processing means for performing data processing of records included in each transmission unit data supplied from the database server;
Final reception position information holding means for holding at least the data of the item of the primary key as final reception position information for the record in the last order constituting the transmission unit data supplied from the database server;
After the data supply from the database server based on the search process requested by the first search requesting unit is interrupted, the first receiving position information holding unit holds the first receiving position information based on the first receiving position information. A second search requesting unit that searches only the records that are not supplied from the database server among the search results of the search processing requested by the search requesting unit, and requests the data transmission processing for the search result. Characteristic data processing device.   The second search requesting unit is a search result of the search process requested by the first search requesting unit after data supply from the database server based on the search process requested by the first search requesting unit is interrupted. The search processing is performed only for the records in the order after the record related to the final reception position information held by the final reception position information holding means, and the data transmission processing is requested for the search result. Item 4. The data processing device according to Item 1. Computer
A search process is executed from a database in which data of a plurality of records is stored, and a predetermined number of records are transmitted as one transmission unit in an order based on a primary key composed of one or a plurality of items for a search result by the search process. First search requesting means for requesting a database server for data transmission processing to be transmitted as data;
Data processing means for performing data processing of records included in each transmission unit data supplied from the database server;
Final reception position information holding means for holding at least the data of the item of the primary key as final reception position information for the record in the last order constituting the transmission unit data supplied from the database server;
After the data supply from the database server based on the search process requested by the first search requesting unit is interrupted, the first receiving position information holding unit holds the first receiving position information based on the first receiving position information. Of the search results of the search process requested by the search request means, only the records that are not supplied from the database server are searched, and the search results are made to function as the second search request means for requesting the data transmission process. A data processing program characterized by that.   In a database system comprising one or more database servers and a data processing device that requests the database server to perform a search process, the data processing device according to claim 1 or 2 is applied as the data processing device. Database system.   In a communication control apparatus comprising: a data processing apparatus that acquires data related to communication control from a database server; and a communication control processing unit that performs communication control processing between communication apparatuses using data acquired by the data processing apparatus. A communication control device, wherein the data processing device according to claim 1 or 2 is applied as a processing device.   A network system comprising a plurality of communication devices and a communication control device that performs communication control processing between the communication devices, wherein the communication control device according to claim 5 is applied as the communication control device. .

Images