JP5211180B2 - Database management apparatus, database management program, database management method, database management system - Google Patents

Description

  The present invention relates to a database management apparatus, a database management program, a database management method, and a database management system.

  In recent years, blog services called miniblogs and microblogs have been provided. In such a blog service, a user can post a short sentence with a limited number of characters, and the posted sentences are displayed in chronological order, thereby providing a place for easy communication between users.

  With this type of service, as the number of users increases, the information posted by users increases over time and becomes enormous. For this information management, a key-value database that allows easy database expansion It is known to use (for example, refer nonpatent literature 1).

Ohei Shinpei, “How to get started with Cassandra: Let's move hands to experience NoSQL”, [online], April 27, 2010, [December 11, 2010], Internet <URL: http: // gihyo.jp/dev/serial/01/cassandra/0001?skip>

  However, the key-value database is a database with specifications that store information (hereinafter sometimes referred to as “value”) in association with a plurality of search keys, and extract values by specifying the search keys.

  For example, if a key-value database is managed as a value for a material (for example, an image) set for the posted text or a specific word / phrase included in the posted text, a search is performed by specifying it. There is a problem that it cannot be done.

  Therefore, if the key-value database manages the information on the material set for the posted text and the specific words / phrases contained in the posted text as values, perform a search that specifies the material. I can't.

  The present invention solves the above-described problems of the prior art, and an object of the present invention is to facilitate information on a specific type of information stored in a key-value database or information based on the specific type of information. It is to make it searchable.

  Based on the above object, one aspect (1) of the present invention is a database management apparatus that acquires information managed by a key-value type first database and manages the information by a non-key-value type second database, The first database is a database that accumulates each element of the posted information in association with a plurality of search keys every time it accepts posted information including specific types of information as elements from external devices used by a plurality of users. A collecting means for collecting the specific type of information among the elements of the posted information managed by the first database, and the specific type of information collected by the collecting means or information based on the specific type of information. Storage processing means for storing in the second database, and the second database by the storage processing means The stored information in response to a request from the external device and transmitting means for transmitting to the external device.

  According to the aspect (6) of the present invention, which grasps the aspect (1) from the viewpoint of a computer program, the information managed by the key-value type first database is acquired and managed by the non-key-value type second database. A database management program for causing a computer to execute processing, wherein each time the first database receives posting information including a specific type of information from an external device used by each of a plurality of users, Are stored in association with a plurality of search keys, and the specific type of information is collected from the elements of the posted information managed by the first database, and the collected specific type of information or the specific Information based on the type of information is stored in the second database, and the second database is stored. The information stored in the scan in response to a request from the external device to execute a process of transmitting to the external device to the computer.

  A database that controls the computer according to one aspect of the present invention that captures the above aspect (1) from the viewpoint of a method, acquires information managed by the first database of the key value type, and manages the information by the second database of the non key value type In the management method, the first database associates each element of the posted information with a plurality of search keys each time it receives posted information including a specific type of information from an external device used by each of a plurality of users. And a procedure for collecting the specific type of information managed by the first database and the information based on the collected specific type of information or the specific type of information. Stored in the second database and the information stored in the second database On demand from and a transmission step of transmitting to the external device.

  An aspect (8) of the present invention that captures the above aspect (1) from the viewpoint of a system is a database management including a key value type first database and a database management apparatus having a non-key value type second database. In the system, the first database accumulates each element of the posted information in association with a plurality of search keys each time it accepts posted information including specific types of information as elements from external devices respectively used by a plurality of users. The database management device is a collection unit that collects the specific type of information among the elements of the posted information stored in the first database, and the specific type of information collected by the collection unit or Storage processing means for storing information based on the specific type of information in the second database; The information stored in the second database by serial storage processing means in response to a request from the external device and transmitting means for transmitting to the external device.

  According to the present invention, it is possible to easily search the second database for specific types of information and information based on the specific types of information among elements of post information stored in the key-value type first database. In addition, since it is not necessary to search the first database for each request from the external device, it is possible to prevent the performance of the first database from being degraded.

  In another aspect (2) of the present invention, in any one of the above aspects, the collecting unit collects the specific type of information from the first database.

  According to the present invention, for example, information can be collected from the first database whenever necessary, so that it is not necessary to provide a special configuration for collecting specific types of information.

  According to another aspect (3) of the present invention, in any one of the above aspects, when the specific type of information to be stored in the first database is transmitted from an external device, the collecting unit Collect information.

  According to this invention, since the search from the first database is not required, it is possible to reduce the processing load on the first database and to prevent the performance of the first database from being lowered.

  According to another aspect (4) of the present invention, in any one of the above aspects, the storage processing unit stores information obtained by tabulating the specific type of information in the second database as information based on the specific type of information. Remember.

  According to this invention, since the information (total information) obtained by totaling specific types of information is stored in the second database, the total information can be searched quickly.

  According to another aspect (5) of the present invention, in any one of the above aspects, the post information element is distributed and managed by a plurality of the second databases, and the plurality of second databases includes each of the post information. A plurality of the same information is distributed and stored for the elements, and the collecting means collects the specific type of information from the latest information among the same information.

  According to the present invention, even when the same information is stored in order to improve maintainability, specific types of information can be collected from the latest information, and a request for searching for as new information as possible can be met.

  According to the present invention, it is possible to easily search for information of a specific type or information based on the specific type of information stored in a key-value database.

FIG. 1 is a schematic diagram of a database management method according to an embodiment of the present invention. FIG. 2 is a configuration diagram of a database management system according to an embodiment of the present invention. FIG. 3 is a diagram illustrating an example of a web page displayed on the terminal device. FIG. 4 is a block diagram of the posted information storage server. FIG. 5 is a diagram illustrating an example of the data configuration of the posting information database. FIG. 6 is a block diagram of the posting information index server. FIG. 7 is a diagram illustrating an example of a data configuration of the posting information index database. FIG. 8 is a flowchart illustrating an example (part 1) of processing performed by the posting information storage server. FIG. 9 is a flowchart illustrating an example (part 1) of processing performed by the posting information index server. FIG. 10 is a flowchart illustrating an example (part 2) of the process performed by the posting information index server. FIG. 11 is a block diagram of a posted information storage server according to another embodiment.

  Hereinafter, modes for carrying out the present invention (hereinafter referred to as “embodiments”) will be described in detail with reference to the drawings.

[1. Overview of database management method]
First, an outline of a database management method according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a schematic explanatory diagram of a database management method according to an embodiment of the present invention. In the following, a blog service system that provides services called microblogs and miniblogs with a limited number of posted characters will be described as an example, but the database management method according to the present embodiment is applied to various systems and devices. It is possible.

As shown in FIG. 1A, the blog service system of the present embodiment includes a plurality of post information storage servers 20 _{1 to} 20 _n and a post information index server 30, and includes a plurality of terminal devices 40 _1. Storage and browsing of information posted from 40 _n is possible. In the following, when referring to any or all of the contribution information storage server 20 ₁ to 20 _n are set to post the information storage server 20, also when referring to any or all of the terminal apparatus 40 ₁ to 40 _n The terminal device 40 is used.

The posted information storage server 20 includes a receiving unit 23b, a storage processing unit 23c, and a posted information database 22. The accepting unit 23b accepts information posted from each of the terminal devices 40 ₁ to 40 _n . The storage processing unit 23c accumulates the information received by the receiving unit 23b in the posted information database 22 (corresponding to an example of the first database) as posted information.

  The posted information database 22 is a key-value database, and each element of posted information is stored in association with a plurality of search keys. A key-value type database is an effective database for a system in which accumulated information increases, such as a blog service system, because it is easy to add a database and to distribute and distribute information. In the example shown in FIG. 1A as well, a plurality of posted information storage servers 20 each having a posted information database 22 are arranged.

  Here, an example of the posting information stored in the posting information database 22 will be specifically described. FIG. 1B is a diagram illustrating an example of a data structure of information stored in the posting information database 22. Here, the posted information and the material ID are included in the posted information. The material ID is an ID for identifying an image (hereinafter referred to as a material image) set for the posted text.

  As shown in FIG. 1B, post information is assigned to the column family “post” among the column families belonging to the key space “main”, and “main” and “post” are search keys, respectively. . Each post information is assigned a post ID as a search key, and a search key is also assigned to each element of the post information.

  Specifically, the posted information elements include a posted text and a material ID, and a search key is assigned to each. In the example shown in FIG. 1B, “post text A” is assigned to the search key “post text” as an element of post information assigned with the search key “post ID” being “100”, and the search key “ “200” is assigned to “Material ID” and stored.

As a post information element from the post information database 22, for example, when searching for a post sentence with a post ID “100”, the search formula shown in (1) below is used, and when searching for a material ID with a post ID “101” The search formula is as shown in (2) below.
Search expression = [main] [post] [100] [post text] (1)
Search expression = [main] [post] [101] [material ID] (2)

  As described above, the posting information database 22 is a database that stores the elements of the posting information in association with a plurality of search keys and specifies the search keys to extract the elements of the posting information.

  For example, in the example shown in FIG. 1B, the material ID with the post ID “100” is stored in association with a plurality of search keys [main], [post], [100], and [search ID]. Is done. Then, the search ID “200” with the post ID “100” can be read out by the search formula (2) that specifies these search keys.

  However, it is not possible to perform a search specifying an element of post information. For example, it is not possible to perform a search specified that “material ID” is “200”. This is because “200” is not a search key.

  Therefore, for example, in order to search for the number of posts by material image (by material ID), a search is performed for the number of posts. For example, if post information with post IDs of “100” to “100000” is accumulated, the post ID is incremented by 1 and [main] [post] [100] [material ID] to [main] [post] ] [100,000] [Material ID] are searched using a search formula. As a result, information on the material ID included in each piece of post information accumulated in the post information database 22 is read out. Then, based on the read material ID information, the number of material images (by material ID) is counted.

  As described above, when a search specifying the material image information (material ID) that is an element of the posted information is performed on the posted information database 22, the posted information database 22 is overloaded and the performance is degraded. A separate calculation process is required to count the numbers for each ID.

  In addition, in the key-value database, post information can be acquired collectively in addition to the above-described search for each post. That is, if the search key = [main] [post], all post information can be acquired. However, after extracting the material ID from each posting information, the number for each material ID must be tabulated, and additional calculation processing is required.

Moreover, blog service system of the present embodiment, the plurality of post information storage server 20 ₁ to 20 _n, on which is managed in a distributed plurality of post information, the same post information (original and replica) also dispersed If post information elements are updated, it takes time to update all of the same information. Therefore, when searching, it is necessary to select the latest information with the latest time stamp from the same information, and a load and time are required for the processing.

  As described above, the key-value type database is effective in a blog service system where the accumulated information increases because the database can be easily expanded and the information is distributed and distributed. ) Cannot be easily performed.

  Therefore, in the blog service system of the present embodiment, the posting information index server 30 is provided so that a search specifying the element of the posting information searched with the search key in the posting information database 22 can be performed very easily.

  As shown in FIG. 1A, the posted information index server 30 includes a collecting unit 33d, a storage processing unit 33e, a posted information index database 32 (corresponding to an example of a second database), a receiving unit 33c, have.

  The collecting unit 33d acquires the material ID of each posted information from the posted information database 22. The search from the posted information database 22 is performed by reading the material ID of each posted information using the search key. Note that the material ID of each piece of posted information can be extracted after reading the entire column family [post].

  The storage processing unit 33e counts the number of postings for each material ID based on the information collected by the collecting unit 33d. For example, the storage processing unit 33e stores the number of postings in the posting information index database 32 using the material ID as an index.

  Here, an example of information stored in the posted information index database 32 will be specifically described. FIG. 1C is a diagram illustrating an example of a data structure of information stored in the posted information index database 32.

  As shown in FIG. 1C, the posting information index database 32 stores the number of posts by material ID with the material ID as an index. For example, the number of posts “14” is associated with the material ID “200”, and the number of posts “20” is associated with the material ID “201”.

  When there is a search request specifying the material ID that is an index from the terminal device 40, the receiving unit 33c reads the number of posts of the specified material ID from the posting information index database 32. Then, the information on the number of posts read in this way is transmitted as browsing information to the terminal device 40 that has made a search request.

  As described above, in the blog service system according to the present embodiment, the posting information index server 30 is provided in addition to the posting information storage server 20, and the posting information index database 32 uses the material ID as an index and posts by material ID. The number of cases is accumulated. Therefore, according to a request from the terminal device 40, the number of posts can be easily read using the material ID as an index.

  In the above description, the material image can be set for the posted text, and by allowing the material image to be viewed together with the posted text, the user who is the viewer can quickly grasp the content of the posted text. In addition, since it is possible to perform a search specifying a material image, a more attractive blog service can be provided.

  In the above description, the material ID is described as an example of the post information element collected by the post information index server 30. However, the post information is not limited to this, and the post information element associated with a plurality of keys in the post information database 22 is described. Anything may be used. For example, when a specific word / phrase included in a posted sentence is managed as a value in the post information database 22, the specific word / phrase can be used as an element of post information collected by the post information index server 30.

[2. Specific examples of database management methods]
Next, a specific example of a blog service system to which the database management method of this embodiment is applied will be described with reference to FIGS. As described above, the database management method according to this embodiment can be applied to various systems and apparatuses.

  First, an overall configuration of a blog service system to which the database management method according to the present embodiment is applied will be described. FIG. 2 is a diagram showing a configuration of the blog service system according to the present embodiment.

As shown in FIG. 2, the blog service system S includes an application server 10, a posting information storage server 20 (20 _{1 to} 20 _n ), and a posting information index server 30. These servers are LAN (Local). Connected by a network such as Area Network. The blog service system S is connected to a network 50 such as the Internet via a router 60.

The blog service system S is a system that allows users of the terminal devices 40 (40 ₁ to 40 _n ) to post and browse. Each terminal device 40 can transmit and receive information to and from the blog service system S via the network 50, and the blog service system S accepts requests such as posting and browsing from the user of each terminal device 40.

  For example, a posting request from a user of each terminal device 40 is accepted by the application server 10, information corresponding to the posting request is notified from the application server 10 to the posting information storage server 20, and posting information of the posting information storage server 20 Accumulated in the database 22 as post information.

  The browsing request from the user of each terminal device 40 is accepted by the application server 10 and notified to the posted information storage server 20 and the posted information index server 30 according to the type. The posting information storage server 20 and the posting information index server 30 read information corresponding to the browsing request from the internal database, and notify the terminal device 40 of the browsing request source via the application server 10.

  Each terminal device 40 includes a communication unit 41, a browser 42, an operation unit 43, and a display unit 44, and controls the browser 42 by an operation on the operation unit 43, via the communication unit 41. Various types of data are transmitted to and received from the application server 10. Thereby, the user of each terminal device 40 can perform posting and browsing.

  Here, an example of a web page for making a posting or browsing request in each terminal device 40 will be described. FIG. 3 is a diagram illustrating an example of a web page transmitted from the application server 10 and displayed on the terminal device 40.

  In the blog service system S according to the present embodiment, for example, when the browser 42 accesses the application server 10 by an operation on the operation unit 43, the application server 10 sends information on the TOP page of the blog service to the terminal device 40. Send. Note that the access to the application server 10 includes user ID information corresponding to the user of the terminal device 40, and a TOP page corresponding to the user ID is transmitted from the application server 10.

  The browser 42 displays the TOP page 11 of the blog service on the display unit 44 based on the TOP page information received from the application server 10. FIG. 3A shows an example of the TOP page 11 displayed on the display means 44. On this TOP page 11, “one word posting”, “one word browsing”, and “used material list image” can be selected.

  When “single word posting” on the TOP page 11 is selected by operating the operation means 43, the selection information is transmitted from the browser 42 to the application server 10. In response to the transmission of the selection information, information on the posting page is transmitted from the application server 10 to the terminal device 40, and the posting page 12 for posting is displayed by the browser 42 as shown in FIG. It is displayed on the means 44. The user of the terminal device 40 inputs a posted message in the posted message input box 12a by operating the operation unit 43, and presses the “post” button 12b. As a result, the posted text information is transmitted to the application server 10.

  In response to the transmission of the posted text, the application server 10 transmits information on the material image setting page 13 as shown in FIG. 3C, and the material image setting page 13 is displayed on the display unit 44 by the browser 42. Is displayed. The material image setting page 13 displays material images 17a to 17c, and further includes a check box 13a for selecting any of the material images 17a to 17c.

  The user of the terminal device 40 selects a material image considered to be related to the posted text posted as described above from the material images 17a to 17c. The material image is selected by checking the check box 13a corresponding to the material image by operating the operation means 43 and pressing the setting button 13b.

  Thereby, the material ID of the material image selected by the user is transmitted to the application server 10. Then, the posted text and the material ID transmitted to the application server 10 are transmitted to the posted information storage server 20 as elements of the posted information and stored in the posted information storage server 20. When the posted information is browsed on the terminal device 40, the browsed image is an image in which the material image 17a and the posted text 16 are arranged adjacent to each other as shown in FIG.

  Thus, by setting the material image related to the content of the posted text, it is possible to enjoy the posting, and it becomes easy to grasp the posted content as shown in FIG. Therefore, the blog service system S according to the present embodiment can provide an attractive blog service that is easy for the user to use.

  In addition, in the blog service system S according to the present embodiment, for example, by selecting “used material image list” on the TOP page 11, a search specifying a material image can be performed.

  When the “use material image list” on the TOP page 11 is selected by operating the operation means 43, the selection information is transmitted from the browser 42 to the application server 10. In response to the transmission of the selection information, information on the used material image list page is transmitted from the application server 10 to the terminal device 40.

  Based on the information received from the application server 10, the browser 42 of the terminal device 40 displays a used material image list page 15 showing a list of material images used for posting by the user, as shown in FIG. 42 is displayed on the display means 44. The used material image list page 15 displays a list of material images and the number of items used by the user for posting. In the example shown in FIG. 3E, it is shown that the number of submissions of the material image 17a is 10 and the number of submissions of the material image 17c is 8.

  The use material image list page 15 is generated based on information acquired by the application server 10 requesting the posting information index server 30 and acquiring it. Specifically, the application server 10 requests the post information index server 30 for information corresponding to the used material image list request transmitted from the terminal device 40.

  Further, on the used material image list page 15, a displayed material image can be selected. For example, in the use material image list page 15 shown in FIG. 3E, for example, when the material image 17a or the material information 15a ("Hima ~") is selected by operating the operation means 43, the selection is made from the browser 42. Information is transmitted to the application server 10. In response to the transmission of the selection information, information on the material image detail page is transmitted from the application server 10 to the terminal device 40.

  Based on the information received from the application server 10, the browser 42 of the terminal device 40 displays the material image detail page 16 showing the details of the material image selected by the user by the browser 42 as shown in FIG. It is displayed on the means 44. The material image detail page 16 displays the number of posts of the selected material image, information on the poster, and the like. In the example shown in FIG. 3F, the total number of postings of the material image 17a is 82176, and the users who are first to third among the many postings are Mr. XX (user ID “543261”), It is shown that they are Mr. △△ (user ID “529076”) and □□ (user ID “175433”). In addition, the material image 17a indicates that the number of posts is “234th” among all the material images.

  As described above, in the blog service system S according to the present embodiment, a posting sentence and a corresponding material image can be posted as posting information from the user of the terminal device 40, and a list of material images used by the user for posting Details of the material image and the like can be viewed from the user of the terminal device 40. Regarding the blog service system S configured as described above, the configuration of the posted information storage server 20 and the posted information index server 30 will be described in detail below.

[2.1. Configuration of Post Information Storage Server 20]
FIG. 4 shows the configuration of the posted information storage server 20. As illustrated in FIG. 4, the posted information storage server 20 includes a communication interface (I / F) 21, a posted information database 22, and a control unit 23.

  The communication I / F 21 is an interface for enabling the posting information storage server 20 to be connected to the application server 10 and the posting information index server 30 via a LAN, and includes, for example, an Ethernet (registered trademark) adapter.

The posting information database 22 is a database that accumulates information related to postings from users of the terminal devices 40 ₁ to 40 _n . The posted information database 22 is a key-value database, and each element of posted information is stored in association with a plurality of search keys. As described above, the key-value database is a database suitable for a system in which accumulated information increases, such as the blog service system S, because it is easy to expand the database and distribute information.

  Here, an example of the posting information stored in the posting information database 22 will be specifically described. FIG. 5 is a diagram illustrating an example of the data configuration of the posting information database 22.

  As illustrated in FIG. 5, the posting information is assigned to a column family “post” among the column families belonging to the key space “main”. Each posting information is assigned a posting ID, which is a posting ID, and a search key is also assigned to each element of the posting information.

  Specifically, the posted information elements include a posted sentence, a posting type, a user ID, a posting date and time, a material ID, and the like, each assigned to a search key. In the example shown in FIG. 5, as information related to a post having a post ID “111”, for example, “Nemu Togashi #Hima” as a post sentence, “0” as a post type, “555” as a user ID, and “ “December 28, 2010, 20:14:27” and “543228” as the material ID are stored.

For example, when searching for a material ID having a posting ID “111” as an element of posting information from the posting information database 22, the search formula is as follows.
Search formula = [main] [post] [111] [material ID]

  Returning to the description of FIG. 4, the description of the configuration of the posted information storage server 20 is continued. The control unit 23 is a control unit that performs overall control of the posted information storage server 20. The control unit 23 further includes a receiving unit 23a, a receiving unit 23b, a storage processing unit 23c, an acquiring unit 23d, and a transmitting unit 23e.

  4 shows a functional configuration of the control unit 23. The control unit 23 is physically composed of a CPU (Central Processing Unit), a flash memory, and a RAM (Random Access Memory). ). The CPU reads out the program from the flash memory and executes it using the RAM as a work area, thereby functioning as the above-described means 23a to 23e.

  The receiving unit 23 a receives post information transmitted from the application server 10 and a search request signal transmitted from the post information index server 30 via the communication I / F 21.

  The accepting unit 23b accepts the information received by the receiving unit 23a. For example, when posting information from the application server 10 is received by the receiving unit 23a, the receiving unit 23b determines whether or not the posting information is within a range to be accumulated, and is within a range to be accumulated. If it is determined, the posted information is output to the storage processing unit 23c. In this case, if the posted information is original information, a replica (duplicate) of the posted information is transmitted from the transmitting unit 23e to the other posted information storage server 20. Similarly, when it is determined that the posted information is not in the range to be accumulated, the accepting unit 23b causes the transmitting unit 23e to transmit to the other posted information accumulation server 20 in which the posted information is to be accumulated.

  In addition, when the search request signal is received from the posting information index server 30 by the receiving unit 23a, the reception unit 23b outputs the search content information included in the search request signal to the acquisition unit 23d.

  The storage processing unit 23c associates the posting ID with the posting information received by the receiving unit 23b and accumulates it in the posting information database 22.

  The acquisition unit 23d acquires information corresponding to the search content (hereinafter referred to as search result information) with reference to the posting information database 22 based on the search content information output from the reception unit 23b. As described above, the posting information database 22 is a key-value database, and the acquisition unit 23d acquires information from the posting information database 22 using a search key.

  For example, it is assumed that a request for searching for material ID information related to all post information from the post information index server 30 is included in the search request signal as search contents. In this case, the acquisition means 23d searches from the post information database 22 using the search key from [main] [post] [100] [material ID] to [main] [post] [100000] [material ID]. Obtain material ID information. It is assumed that the posting information database 22 stores posting information with posting IDs “100” to “100000”.

  Further, for example, it is assumed that a request for searching for material ID information related to post information for which a range is specified is included in the search request signal as search contents from the post information index server 30. In this case, the acquiring unit 23d searches for the material ID of the posted information included in the designated range by using the search key, and acquires the material ID information from the posted information database 22. For example, when the post ID is in a range from “10001” to “30000”, the acquisition unit 23d selects [main] [post] [10001] [material ID] to [main] [post] [30000] [ Search up to “Material ID” with the search key.

  The transmission unit 23e transmits, for example, the search result information acquired by the acquisition unit 23d to the posting information index server 30.

[2.2. Configuration of Post Information Index Server 30]
FIG. 6 shows the configuration of the posting information index server 30. As illustrated in FIG. 6, the posting information index server 30 includes a communication interface (I / F) 31, a posting information index database 32, and a control unit 33.

  The communication I / F 31 is an interface for enabling the posting information index server 30 to be connected to the application server 10 and the posting information storage server 20 via a LAN, and includes, for example, an Ethernet (registered trademark) adapter.

  The posted information index database 32 is a storage unit that stores information related to the material ID of posted information acquired from the posted information storage server 20 by the control unit 33. The posted information index database 32 is, for example, a relational database, and commands according to, for example, SQL (Structured Query Language) are used to store, read, and update cache data.

  Here, the information managed in the posting information index database 32 will be specifically described with reference to the drawings. FIG. 7 is a diagram illustrating an example of a data configuration of the posting information index database 32.

  In the posting information index database 32, the material ID total information includes the material ID total information for each user ID, the material ID total information for the total post information, and the material ID total information for a plurality or all user ID units. Accumulated.

  Specifically, in the posting information index database 32, as shown in FIG. 7A, the material ID and the number of postings are stored in association with each user ID. For example, in the example shown in FIG. 7A, for the user whose user ID is “555”, the number of posts of the material ID “543228” is “10” and the material ID is “287630”. This indicates that the number of posts of “24” is “24”.

  Further, in the posted information index database 32, as shown in FIG. 7B, accumulated information of material IDs for the total posted information is accumulated. In the example shown in FIG. 7B, for the material ID “543228”, the number of postings is “82176” and the number of postings is the highest among the users (poster (first place)) to the third (poster (poster ( It is shown that the user IDs up to the third place)) are “543261”, “529076”, and “175433”. Further, the material image of the material ID “543228” indicates that the number of posts is the “234” th most among all the material IDs.

  Returning to FIG. 6, the description of the configuration of the posting information index server 30 will be continued. The control unit 33 is a control unit that performs overall control of the posting information index server 30. The control unit 33 further includes transmission means 33a, reception means 33b, reception means 33c, collection means 33d, and storage processing means 33e.

  6 shows a functional configuration of the control unit 33. The control unit 33 is physically configured by a CPU (Central Processing Unit), a flash memory, and a RAM (Random Access Memory). ). The CPU reads the database management program from the flash memory and executes it using the RAM as a work area, thereby functioning as the above-described means 33a to 33e.

  The transmission unit 33a transmits, for example, aggregate information of material IDs, which are information stored in the posted information index database 32, to the application server 10 via the communication I / F 31, and also transmits a search request signal for requesting a search. It transmits to the posting information storage server 20 via the communication I / F 31.

  The receiving unit 33b receives, for example, the information request signal transmitted from the application server 10 and the search result information transmitted from the posting information storage server 20 via the communication I / F 31. The information request signal is a signal for making an information transmission request from the application server 10 to the posted information index server 30 and includes information on search contents.

  The accepting means 33c accepts the information request signal received by the receiving means 33b, reads information corresponding to the information request signal from the posted information index database 32, and causes the application server 10 to transmit from the sending means 33a.

  For example, when information is accumulated in the posted information index database 32 as shown in FIG. 7A, the information of the search content included in the information request signal is used by the user with the user ID “555”. It is assumed that the information is a material image. In this case, the accepting unit 33c specifies the user ID “555” as an index, reads all the material IDs and the number of posts associated with the user ID “555” from the post information index database 32, and transmits the search unit 13a as a search result. To the application server 10.

  The collection unit 33d causes the transmission unit 13a to transmit a search request signal including information about the search contents to the posted information storage server 20. The posting information storage server 20 refers to the posting information database 22 based on the information of the search contents included in the search request signal, acquires information according to the search contents, and transmits the information to the posting information index server 30.

  When the collecting means 33d collects the information of the search contents included in the information request signal from the posted information storage server 20, the collecting means 33d performs a counting process and outputs it to the storage processing means 33e, and stores the result of the counting process in the posted information index database 32. Let

  For example, when the information of the search content included in the search request signal is information of a request for searching for material ID information related to all post information, the acquisition unit 23d of the post information storage server 20 uses the material ID of all post IDs. Is transmitted to the posting information index server 30.

  When the collecting unit 33d collects material ID information from the posting information storage server 20 regarding all posting information (for example, posting information with posting IDs “100” to “100000”), the collecting unit 33d collects the information for each user ID. The number by material ID is totaled. For example, it is assumed that “833457”, “257638”, “833457”, “83457”, and “257638” are transmitted from the posted information storage server 20 as material IDs included in the posted information having the user ID “283” as an element. In this case, the collection unit 33d sets the number of posts of the material ID “83457” to “3” and the number of posts of the material ID “257638” to “2”. This is the totaling result of the collecting means 33d.

  In addition, when the collecting unit 33d collects material ID information related to all posted information, the collecting means 33d can collect material IDs only for posted information newly stored in the posted information storage server 20 next time. Thereby, compared with the case where the information of material ID regarding all the contribution information is collected and totaled every time, the processing load, the communication load, etc. of the contribution information storage server 20 and the contribution information index server 30 can be reduced.

  For example, when the collection unit 33d collects and totals material IDs up to a post ID “100000”, the collection unit 33d accumulates post ID information related to post information with a post ID greater than “100000”. Request to server 20. Then, the collecting unit 33d updates the aggregate information stored in the posted information index database 32 based on information transmitted from the posted information storage server 20 in response to this request.

  For example, as a totaling result regarding the user ID “283”, as described above, the number of posts of the material ID “83457” is “3” and the number of posts of the material ID “257638” is “2”. As the information of the material ID of the posted information related to the user ID “283” of the posted information with the posted ID larger than “100,000”, “833457”, “8345747”, and “257638” are the posted information storage server 20. Is sent from. In this case, the collection unit 33d updates the count result regarding the user ID “283” with the number of posts of the material ID “833457” being “5” and the number of posts of the material ID “257638” to “3”.

  The storage processing unit 33e stores the aggregation results of the collecting unit 33d in the posting information index database 32. As a result, for example, aggregate information of material IDs as shown in FIGS. 7A and 7B is accumulated in the posting information index database 32.

  A specific operation of the posted information storage server 20 configured as described above will be described with reference to FIG. FIG. 8 is a flowchart showing a processing procedure executed by the control unit 23 of the posted information storage server 20.

  This process is a process that is repeatedly performed. The CPU that constitutes the control unit 23 performs the process based on the program stored in the flash memory, and it is easy to understand the relationship with the functional blocks shown in FIG. Therefore, for the sake of convenience, description will be made using the expression that each functional block is executed.

  As shown in FIG. 8, in the posted information storage server 20, the accepting unit 23b determines whether or not posted information has been received from the application server 10 (step S10). In this process, if it is determined that the posted information has been received (step S10, Yes), the receiving unit 23b determines whether the received posted information is within a range to be accumulated (step S11).

  When the posted information received by the receiving unit 23b is within the range to be accumulated (step S11, Yes), the storage processing unit 23c stores the posted ID in association with the received posted information in the posted information database 22 (step S11). S12). In addition, the transmission unit 23e transmits a post information replica (duplicate) to another post information storage server 20 for storage (step S13).

  In step S11, when the posted information received by the receiving unit 23b is not within the range to be stored (step S11, No), the transmitting unit 23e sends the posted information received by the receiving unit 23b to another posted information storage server 20. Transfer (step S14).

  If it is determined in step S10 that post information has not been received (step S10, No), the receiving unit 23b determines whether a search request signal has been received from the post information index server 30 (step S15). In this process, if it is determined that the search request signal is received (step S15, Yes), the acquisition unit 23d acquires information (search result information) according to the search content included in the search request signal from the posted information database 22. Then, the transmission unit 23e transmits the search result information to the posting information index server 30 (step S16).

  Further, the transmission unit 23e requests the other posting information storage server 20 having the replica of the posting information searched in step S16 to transmit the replica to the posting information index server 30, and the other posting information storage server 20 Is sent to the posting information index server 30 (step S17).

  On the other hand, if it is determined in step S15 that a search request signal has not been received from the posting information index server 30 (step S15, No), the control unit 23 performs processing according to other requests (step S18). When the processes of steps S13, S14, S17, and S18 are finished, the process shown in FIG. 8 is finished.

  Next, a specific operation of the posting information index server 30 will be described with reference to FIG. FIG. 9 is a flowchart showing a processing procedure executed by the control unit 33 of the posting information index server 30.

  This process is a process that is repeatedly performed. The CPU constituting the control unit 33 performs the process based on the database management program stored in the flash memory, and the relationship with the functional blocks shown in FIG. For the sake of simplicity, description will be made using the expression that each functional block is executed for convenience.

  As shown in FIG. 9, in the posting information index server 30, the collection unit 33d determines whether or not an information collection opportunity has been triggered (step S20). For example, the collection unit 33d determines that the information collection trigger has been triggered during a predetermined period (for example, every 5 seconds).

  In this process, if it is determined that an information collection opportunity has been reached (Yes in step S20), the collection unit 33d displays information on a material ID for generating or updating the aggregate information determined to have been an information collection opportunity. 20 is requested (step S21). When collecting the material ID information from the posting information storage server 20, the collecting unit 33d aggregates the material ID information and generates or updates the total information (step S22). Then, the storage processing unit 33e accumulates the total information generated or updated by the collecting unit in the posted information index database 32 (step S23).

  As described above, the post information storage server 20 generates and distributes replicas (duplicates), and the replicas are also transmitted to the post information index server 30. The collecting unit 33d selects the latest information from the same information including the replica and uses it for the aggregation process. A time stamp (time information) is added to the information managed by the posting information index server 30, and the collecting unit 33d refers to the time stamp to determine whether the information is the latest information.

  If it is determined in step S20 that the information collection trigger has not been reached (step S20, No), the receiving unit 33c determines whether an information request signal has been received from the application server 10 (step S24). In this process, if it is determined that the information request signal is received (step S24, Yes), the receiving unit 33c reads the aggregate information corresponding to the information request signal from the posted information index database 32 and transmits it to the application server 10 by the transmitting unit 33a. (Step S25).

  On the other hand, if it is determined in step S24 that an information request signal has not been received (step S24, No), the control unit 23 performs processing according to other requests (step S26). When the processes of steps S23, S25, and S26 are finished, the process shown in FIG. 9 is finished.

  As described above, the control unit 33 of the posted information index server 30 according to the present embodiment acquires information managed by the posted information database 22 (corresponding to an example of the first database) as a database management apparatus, and acquires the posted information index database. 32 (corresponding to an example of the second database).

  The post information database 22 is a database that stores post information elements in association with a plurality of search keys each time it accepts post information including post text and image information set for the post text as elements. is there.

  Then, the control unit 33 of the posting information index server 30 collects the material ID information among the elements of the posting information managed by the posting information database 22, and calculates the total information obtained by collecting the collected material ID information. Store in the index database 32. And the control part 33 transmits the total information memorize | stored in the contribution information index database 32 to an external device according to the request | requirement from an external device.

  Accordingly, it is possible to easily search the posting information index database 32 for aggregate information such as the number of postings for each material image set in the posting text among the posting information stored in the key-value type posting information database 22. In addition, since it is not necessary to search the posting information database 22 for each request from the external device, it is possible to prevent the posting information database 22 from being lowered.

  In the above description, the aggregated information of the material IDs has been described as an example of the information collected and accumulated by the posted information index server 30. You may make it accumulate | store in the database 32 by the memory | storage process means 33e. Further, the material ID information may be accumulated in the posting information index database 32 by the storage processing unit 33e together with the total result. In this case, the application server 10 can collect material ID information from the posting information index server 30, aggregate the information, and transmit it to the terminal device 40.

[3. Other Embodiments]
In the above description, the control unit 33 of the posting information index server 30 searches and collects information on material IDs, which are elements of posting information managed by the posting information database 22, from the posting information database 22. The method is not limited to this. This will be specifically described below.

(Other embodiment 1)
First, another embodiment 1 will be described. In this embodiment, when the posting information storage server 20 receives posting information, the posting information is stored in the posting information database 22 and transmitted to the posting information index server 30. Thereby, since the search from the posting information database 22 is not required, the processing load on the posting information database 22 can be reduced, and the performance of the posting information database 22 can be prevented from being lowered. Further, the posting information index database 32 can be updated in synchronization with the posting information stored in the posting information database 22 in real time.

  Specifically, when the receiving unit 23b of the posted information storage server 20 determines that the posted information received by the receiving unit 23a is within the range to be stored, the posted information is transmitted to the posted information index server 30 by the transmitting unit 23e. To send from. Thereby, the posted information stored in the posted information database 22 can be transmitted to the posted information index server 30 without searching the posted information database 22. The post information transmission process to the post information index server 30 is performed, for example, before the process of step S12 shown in FIG.

  The accepting means 33c of the posted information index server 30 accepts the posted information transmitted from the posted information storage server 20, and the collecting means 33d performs a material ID counting process. Processing in this case will be described with reference to FIG. FIG. 10 is a flowchart showing another processing procedure executed by the control unit 33 of the posting information index server 30. Note that the processing of steps S23 to S26 shown in FIG. 10 is the same as the processing of S23 to S26 shown in FIG.

  In the posted information index server 30, the receiving unit 33c determines whether or not the posted information transmitted from the posted information storage server 20 has been received (step S20 '). In this process, if the receiving unit 33c determines that the posting information has been received (step S20 ′, Yes), the collecting unit 33d extracts the user ID and the material ID from the posting information received by the receiving unit 33c, and these information The number of posts associated with is read from the post information index database 32, and 1 is added to update the number of posts (step S22 '). Then, the storage processing unit 33e updates the posting information index database 32 with the number of postings updated by the collecting unit 33d (step S23).

  For example, the accepted posting information is posting information from the user ID “555”, the material ID is “543228”, and the user ID “555” is used by the user as shown in FIG. It is assumed that the number of postings for each material ID is accumulated in the posting information index database 32. In this case, the collection unit 33d adds 1 to the number of postings of the material ID associated with the user ID “555” and updates the count result as “11”. Then, the storage processing unit 33e updates the posting information index database 32 with the aggregation result updated by the collecting unit 33d.

  If there is no number of posts associated with the user ID and material ID extracted from the posted information received by the receiving unit 33c, the collecting unit 33d newly sets the number of posted material ID posts to “1”. Total information is generated, and the post information index database 32 is updated by the storage processing means 33e.

  In addition, since the post information index server 30 does not need information such as a post text and a post type, for example, it is desirable to acquire only a user ID and a material ID and perform a tabulation process based on these information. As a result, the posting information storage server 20 does not need to transmit all the posting information, and only needs to transmit the user ID and the material ID. Therefore, the communication load between the posting information storage server 20 and the posting information index server 30 Is reduced.

  In addition, when the posting information index server 30 receives the posting information transmitted from the posting information accumulation server 20, the collection unit 33d performs the process of counting the material ID, but the received posting information is not illustrated. When the information request signal is received from the application server 10 by being stored in the storage unit, the collection unit 33d may perform the material ID aggregation process.

(Other embodiment 2)
Next, another embodiment 2 will be described. In this embodiment, the post information storage server 20 ₁ to 20 _n, generates search data from stored in posting information database 22 data, so as to obtain the search data by post information index server 30.

  Hereinafter, the configuration of the posted information storage server according to another embodiment 2 will be described. FIG. 11 is a block diagram of the posted information storage server 20 according to another embodiment 2. In the following, functions added to the posted information storage server 20 shown in FIG. 2 will be mainly described.

  As shown in FIG. 11, the post information storage server 20 according to another embodiment 2 includes a storage unit 24, and the acquisition unit 23 d ′ is a search function in addition to the function of the acquisition unit 23 d illustrated in FIG. 2. It has a function of performing generation data generation processing.

  For example, the acquisition unit 23 d ′ of each posting information storage server 20 executes a process of dumping data in the posting information database 22 to the storage unit 24. Specifically, the acquisition unit 23 d ′ reads all the posted information accumulated in the posted information database 22 and stores it in the storage unit 24 as a dump file. The acquisition unit 23d ′ performs this dump process periodically (for example, once a day).

  In this way, the dump file stored in the storage unit 24 includes both an original (master) and a replica (replica). Therefore, the acquisition unit 23d ′ extracts only the original from the dump file stored in the storage unit 24 and stores it in the storage unit 24 as sub data for search.

Further, since the original post information is distributed and stored in the post information database 22 of the plurality of post information storage servers 20 _{1 to} 20 _n , one post information storage server 20 (hereinafter referred to as “specific”) is specified in advance. Acquisition means 23d 'collects search sub-data from all other post information storage servers 20. Then, the acquisition unit 23d ′ of the specific posting information storage server 20 merges the generated search subdata and the collected search subdata to generate search data.

Specifically, the acquisition unit 23d ′ of the specific post information storage server 20 sends a search sub-data transmission request to all other post information storage servers 20 from the transmission unit 23e. Then, the acquisition unit 23d ′ of the specific post information storage server 20 stores, in the storage unit 24, the search sub-data transmitted from the other post information storage server 20 and received by the reception unit 23b in response to this transmission request. By doing in this way, the search sub-data of the posting information storage servers 20 _{1 to} 20 _n is stored in the storage unit 24 of the specific posting information storage server 20. Then, the acquisition unit 23d ′ of the specific post information storage server 20 merges (combines) the search sub data of the post information storage servers 20 _{1 to} 20 _n stored in the storage unit 24 to generate search data. . The acquisition unit 23 d ′ transmits the search data generated in this way to the posting information index server 30.

  Further, the acquisition unit 23d ′ of the posted information storage server 20 performs the search sub-data difference detection process at a time interval shorter than the dump process. Specifically, the acquisition unit 23d ′ of each posting information storage server 20 reads the posting information newly added to the posting information database 22 periodically (for example, every 10 minutes) using the search key and reads the posting. Only the original is extracted from the information and stored in the storage unit 24 as difference data. Then, as in the case of the search sub-data, the acquisition unit 23d ′ of the specific post information storage server 20 collects the difference data from all the other post information storage servers 20, and generates the difference data and the generated difference. The difference data for search is generated by merging the data. The acquisition unit 23 d ′ transmits the search difference data generated in this way to the posting information index server 30.

  The posting information index server 30 acquires search data and search difference data from the specific posting information storage server 20, and extracts the material ID of the posting information from these data. Specifically, the collection unit 33d of the posting information index server 30 stores search data and search difference data in an internal storage unit (not shown), and when an information request signal is received from the application server 10, The material ID is aggregated. Note that the collection unit 33d merges (combines) the search difference data with the search data and updates the search data stored in the internal storage unit. Further, when newly collecting search data, the collecting unit 33d deletes old search data stored in the internal storage unit.

  Here, the collection unit 33d performs the material ID aggregation processing when the information request signal is received from the application server 10, but the received data when the search data or the search difference data is received. On the basis of the above, the material ID may be aggregated by the collecting means 33d.

  In this way, each posting information storage server 20 is provided with a function for performing batch processing, so that it is not necessary to perform processing every time new posting information is generated, which is effective when there is no need to synchronize in real time. It is.

In the above description, the search data and the search differential data are transmitted from the specific post information storage server 20, but the specific post information storage server 20 is responded to the search request signal from the post information index server 30. However, the search result information may be extracted from the search data or the search difference data and transmitted to the posting information index server 30. In this case, when the search request signal from the posting information index server 30 is received by the receiving unit 23a, the receiving unit 23b outputs the search content information included in the search request signal to the acquiring unit 23d ′. When the information on the search content output from the accepting unit 23b is a request from the posted information index server 30, the acquiring unit 23d 'searches for the search data or the search differential data stored in the storage unit 24 instead of the posted information database 22. Information corresponding to the search content (search result information) is acquired from. And then. The transmission unit 23e transmits the search result information acquired by the acquisition unit 23d to the posting information index server 30. Not only specific post information storage server 20, all of the posts the information storage server 20 ₁ to 20 _n may be generated search data and search for the differential data.

  In the above-described embodiment (including the other embodiments 1 and 2), the application server 10 requests and obtains the aggregation result from the posting information index server 30. You may make it request | require and acquire a total result etc. to the contribution information index server 30. FIG.

  In the above-described embodiment (including other embodiments 1 and 2), the material ID is described as an example of the post information element collected by the post information index server 30. However, the present invention is not limited to this. Any element of posted information associated with a plurality of keys in the information database 22 may be used. For example, when a specific word / phrase included in a posted sentence is managed as a value in the post information database 22, the specific word / phrase can be used as an element of post information collected by the post information index server 30.

  As described above, some of the embodiments of the present invention have been described in detail with reference to the drawings. However, these are merely examples, and various embodiments can be made based on the knowledge of those skilled in the art including the aspects described in the section of the disclosure of the invention. The present invention can be implemented in other forms that have been modified or improved.

10 Application server 20 (20 ₁ ~20 _n) posts the information storage server 30 posts the information index server 40 (40 ₁ ~40 _n) terminals 33a transmitting means 33b receiving means 33c receiving unit 33d collecting means 33e storage processing unit

Claims (7)

A database management device that acquires information managed by a key-value type first database and manages the information by a non-key-value type second database,
From an external device using multiple users respectively, the posted information accepting means for accepting post information including information of a particular type to the element,
Each time the posted information is received by the posted information receiving means, a first storage processing means for storing elements of the received posted information in association with a plurality of search keys in the first database,
Each time the posted information is received by the posted information receiving means, the second database stores information obtained by counting the number of the specific type of information based on the specific type of information included in the received posted information. Second storage processing means;
A database management apparatus comprising: a transmission unit configured to transmit information stored in the second database by the second storage processing unit to the external device in response to a request from the external device. The posted information includes posted text information and image ID information,
The database management apparatus according to claim 1, wherein the specific type of information is ID information of the image . The posting information receiving means
From among a plurality of images according to claim 2, characterized in that <br/> accepting the ID information of the image selected by the user of the external device from the external device as the ID information of the image included in the post information Database management device. The post information element is:
Distributedly managed by a plurality of the second databases,
The plurality of second databases are:
Wherein each element of the post information, the database management apparatus according to any one of claims 1 to 3, characterized in <br/> be distributed to a plurality store the same information. A database management program for causing a computer to execute processing for acquiring information managed by a key-value type first database and managing the information by a non-key-value type second database,
From an external device using multiple users respectively, receiving post information including information of a particular type to the element,
Each time the post information is received , the received post information elements are stored in the first database in association with a plurality of search keys, respectively .
Each time the post information is received, information obtained by tabulating the number of the specific type of information based on the specific type of information included in the received post information is stored in the second database,
A database management program for causing a computer to execute processing for transmitting information stored in the second database to the external device in response to a request from the external device. A database management method for controlling a computer to acquire information managed by a key-value type first database and managing it by a non-key-value type second database,
From an external device using multiple users respectively, the reception process for receiving post information including information of a particular type to the element,
A first storage processing procedure for storing each element of the received posting information in association with a plurality of search keys in the first database each time the posting information is received by the receiving procedure ;
A second storage that stores information obtained by counting the number of information of a specific type based on the specific type of information included in the received post information in the second database every time the post information is received by the reception procedure. Processing procedure and
A database management method comprising: a transmission procedure for transmitting information stored in the second database to the external device in response to a request from the external device. In a database management system comprising a first database of key value type and a database management device having a second database of non-key value type,
The database management device includes:
Post information receiving means for receiving post information including specific types of information as elements from external devices used by a plurality of users ,
Each time the posted information is received by the posted information receiving means, a first storage processing means for storing elements of the received posted information in association with a plurality of search keys in the first database,
Each time the posted information is received by the posted information receiving means, the second database stores information obtained by counting the number of the specific type of information based on the specific type of information included in the received posted information. Second storage processing means;
A database management system comprising: a transmission unit configured to transmit information stored in the second database by the storage processing unit to the external device in response to a request from the external device.

Images