JP4512184B2 - Information retrieval system - Google Patents

Description

  The present invention relates to a WEB system, and more particularly, to a server that searches a user who uses a WEB page or information displayed on the WEB page in response to a request from a client, and an information search method in the WEB system.

It is routinely performed that a client accesses a WEB site via a network such as the Internet and acquires information on a WEB page.
Usually, a remote site, which is a WEB site, includes a WEB server having a function of acquiring information in response to a request from a client and transmitting the information, and a DB server that obtains target information by searching a database (DB). In accordance with the content of an HTTP (Hyper Text Transfer Protocol) request transmitted from the client, the DB server obtains target information, creates a WEB page on the WEB server, and responds to the client.

In order to promptly send a target WEB page as a response to a request from a client, it is preferable that acquisition of information on the WEB site, for example, DB search by a DB server is performed at high speed. Conventionally, various proposals have been proposed for this purpose.
For example, Patent Document 1 (Japanese Patent Laid-Open No. 2001-236357) discloses a Web search device that performs a search for a search word designated by a client, even when a desired keyword is not known, effectively searches a Web page. In order to make it possible, a keyword corresponding to the search word and a keyword managed in one or both of the front and rear of the keyword are acquired.

JP 2001-236357 A

  Normally, the information to be searched by the Web server includes various information such as customer name, telephone number, address, service content, etc., and the time required for searching for these information varies. The Web server is kept waiting until the retrieval result of the information that takes the longest time is obtained, so that sending the retrieval result to the client is delayed.

  Further, when providing a service to a user through a Web system, it is necessary to authenticate the user at the start of the service. Usually, the authentication process is performed by collating the ID and password transmitted from the user with these pieces of information registered in the DB in advance. However, if the number of users increases, the amount of information in the DB also becomes enormous, so it takes time to search the DB, which in turn increases the processing time required for authentication, leading to service degradation.

Furthermore, in a Web system that uses a business system that manages information on a large number of users, a user profile, for example, a database that stores basic information such as names, addresses, and telephone numbers, and authority, is searched for users. However, when the number of users increases, there is a problem that it takes time to search for the information.
Thus, in providing a service using a WEB system or in business processing, taking time to acquire necessary information leads to a decrease in service and business efficiency.

  An object of the present invention is to provide a WEB system and an information search processing method in which the time for searching information is shortened.

The present invention relates to an information search processing method in a WEB system for searching for information in a search server in relation to a request from a client, the step of analyzing an HTTP request from the client, and the result of this analysis, in-memory A step of searching information stored in the DB, a step of generating HTML format data for the information obtained as a result of the search, a step of transmitting HTML format data to the client according to HTTP, Is a search processing method.
For example, HTML format data is compressed and then transmitted to the client according to HTTP.
In one example, the HTTP request from the client is analyzed, the HTTP request is transmitted to the WEB server, the response is received, and the in-memory DB is searched according to the content of the response.
Preferably, the method further comprises a step of registering information acquired from a certain DB as a DB replica in the in-memory DB by a DB server connected to the in-memory DB search server, and the information in the in-memory DB is a certain DB It is updated according to the update of the information.
In a preferred example, the in-memory DB stores information of a type that takes time to search. For example, in-memory DB stores user information including any of IDs, user names, telephone numbers, and the like related to users of the WEB system. When accessing a WEB server from a client, in-memory DB DB is searched and these information is acquired.
Further, by analyzing the HTTP request, information not registered in the in-memory DB is acquired from the DB server via the WEB server.
The present invention is also understood as a program for a search server that performs processing for executing the above method.

  In the WEB system according to the present invention, in the WEB system having a search server that searches for information in relation to a request from the client, the information obtained by searching the DB by the DB server in response to the request from the client. From an in-memory DB that stores certain information as a replica, an HTTP request analysis unit that analyzes an HTTP request transmitted from a client, and transmits an HTTP request to the WEB server as necessary, and an HTTP request analysis unit The in-memory DB search request processing unit that issues a search request to the in-memory DB in accordance with the request and analyzes the search result information obtained from the in-memory DB, and the HTTP response analysis unit that analyzes the information acquired from the WEB server And information retrieved from the in-memory DB and / or obtained from the WEB server A response HTML generation unit that edits the information into an HTML format and a content compression unit that compresses the information from the response HTML generation unit according to a certain rule in order to transmit the information to the client. Alternatively, it is a WEB system in which information obtained from a WEB server is compressed by a content compression unit and transmitted to the client.

  The present invention can also be understood as an information processing method in a WEB system that acquires information from a WEB server in association with a request from a client and transmits the information as a response to the client. That is, an information processing method according to the present invention includes a step of analyzing a request from a client, a first mode of acquiring information stored in the in-memory DB as a result of analysis and transmitting a response to the client, As a result, the second mode in which information is acquired from the WEB server and information is acquired from the in-memory DB, and a response is transmitted to the client. As a result of the analysis, the information is not acquired from the in-memory DB, and the WEB server And a third mode in which only the acquired information is sent as a response to the client.

  According to the present invention, information can be searched at high speed with a search server in a WEB system. More specifically, information that would take time to search for information using the DB server is registered in the in-memory DB in advance, and this in-memory DB is searched in response to a request from the client. The time for obtaining information at the remote site can be reduced as a whole.

Hereinafter, an embodiment of the present invention will be specifically described with reference to the drawings.
FIG. 1 is a diagram showing an example of a WEB system to which an embodiment of the present invention is applied.
The remote site 1 and the local site 2 are connected via a network 3 such as the Internet or a WAN (Wide Area Network), and can communicate with each other by HTTP.
The local site 2 is a client (hereinafter, the local site 2 is referred to as a client 2), and usually includes a plurality of PCs 20 (shown as one PC), and the PC 20 has a browser 21. The browser 21 is application software for browsing the WEB page, acquires HTML format files, image files, music files, etc. via the network 3, analyzes the layout, plays them, and displays them on the PC display To do. In general, a local proxy is interposed between the browser 21 and the network 3 as software or a computer for connecting to the network 3 and communicating information, but is omitted in the drawing.

On the other hand, the remote site 1 includes a DB search server 10, a WEB server 16, and a DB server 18. The DB search server 10 has a proxy processing function 12 and an in-memory DB 14. The in-memory DB 14 is a DB formed in a semiconductor memory, and can read and write information at a higher speed than an external storage device such as a disk device accessed by the DB server 18 as a search target.
The configuration or processing operation of the DB search server 10 and the processing related to the search by the DB server 18 are characteristic of the embodiment of the present invention and will be described in detail with reference to FIG.

  FIG. 2 shows a module configuration of the DB search server 10 at a remote site of the WEB system.

The proxy processing function 12 includes a basic processing unit 121, an HTTP request analysis unit 122, an in-memory DB search request processing unit 123, a response HTML generation unit 124, a content compression unit 125, and an HTTP response analysis unit 126. The functions of these units are realized by executing a program having these functions.
The basic processing unit 121 performs basic processing such as initial setting processing and child thread creation.
The HTTP request analysis unit 122 analyzes the HTTP request transmitted from the client 2. As a result of the analysis, if a search of the in-memory DB 12 is necessary according to the HTTP request content, a request is sent to the in-memory DB search request processing unit 123. When processing for the WEB server 16 is necessary, an HTTP request is issued to the WEB server 16. The process for the WEB server 16 is a process for acquiring a necessary object from the WEB server 16, for example.

The in-memory DB search request processing unit 123 issues a search request to the in-memory DB 14 and analyzes the search result from the in-memory DB 14.
The response HTML generation unit 124 edits the information retrieved from the in-memory DB and the information acquired from the WEB server 16 into an HTML format for transmission to the client 2.
The content compression unit 125 compresses WEB page information to be transmitted to the client 2, that is, information including a plurality of normal contents according to a certain rule. The HTTP response analysis unit 126 analyzes information acquired from the WEB server 16.

  The in-memory DB 14 is a processing unit 141 that performs an in-memory DB search process as a search engine, a DB replica unit 142 that is a storage unit for storing information, a process for registering a DB replica, and a registered A DB replica processing unit 143 that performs processing for updating information of the DB replica is configured. The processing operation by the DB replica processing unit 143 will be described later with reference to FIG.

  The DB replica unit 142 originally stores information acquired from a certain DB by the execution of the DB server 18 as a replica, that is, a replica. Further, the information stored in the DB replica unit 142 is updated in accordance with the update of the information of a certain DB that is a replica.

  As the information stored in the DB replica unit 142, it is preferable to store information of a type that increases the amount of information and takes time to search. This is because information that does not require time for searching may be acquired by searching the corresponding DB by the DB server 18. There are various types of information requested from the client 2 by the user of the WEB system. By obtaining information that takes time to search from the DB replica unit 142, the response time to the client can be substantially shortened. it can.

  Information that increases the amount of information and takes time to search includes, for example, information related to users using the WEB system, that is, user ID, user name, telephone number, service content, number of remaining incidents There is information about status, etc. It may be necessary for the user of the WEB server 16 to check whether the user is a valid user. In this case, if information such as the user ID and the user name can be acquired from the in-memory DB 14 at a high speed, the time required for the check can be shortened.

Next, DB replica creation processing will be described with reference to FIGS.
In FIG. 6, in the in-memory DB 14, by executing the DB replica processing unit 143, in the initial processing, the replica table list C is loaded, loaded into the DB replica unit 142, and information to be registered is set (S51). The replica table list C includes a replica condition list D. This replica table list C is stored in a predetermined area of the in-memory DB 14.

As shown in FIG. 9, the replica condition list D includes descriptions of the data extraction interval 81 from the target DB, the replica table list 82, the extraction condition 83, and the extraction source DB 84.
The data extraction interval 81 represents a time interval for acquiring data from the DB to be replicated. In the illustrated example, the data extraction interval 81 is started at 00:00 every day to acquire data.
The replica table list 82 represents the replica table name of the registration destination of the extracted replica. In the illustrated example, a table name in which a telephone number (TEL), a customer name, and a service name are registered is represented by the name “customer information”.

  The extraction condition 83 is represented by a general SQL syntax as illustrated. For example, “TEL”, “customer name”, and “service name” are selected from the “phone number master” that is the extraction source DB. The extraction source DB 84 represents the name of the DB server to be extracted, and is represented as “DBSV01” in the drawing.

  In accordance with the conditions specified by the replica table list C initialized as described above, data acquisition and update processing of the DB replica is performed at a specified time (S52). In this process, data is acquired from the DB specified by the list D under the execution of the DB server 18, and the data of the corresponding DB replica unit 142 is rewritten to update its contents.

Next, an information search processing operation in the WEB system according to the present embodiment will be described.
Prior to a detailed description of the processing operation, an overall overview of the information search process will be described with reference to FIG. In the present embodiment, information retrieval is performed by three types of operations (modes) in accordance with a request from the client 2.

The first mode is a mode in which the in-memory DB 14 is searched for information according to the analysis result of the request from the client 2 and information is acquired and transmitted to the client as indicated by a solid line in FIG. This will be described in detail with reference to FIG.
In the second mode, as indicated by a thin broken line, according to the analysis result of the request from the client 2, the WEB server 16 is accessed to acquire information, and information is also acquired from the in-memory DB 14. This is a mode for transmitting to the client 2. This will be described in detail with reference to FIG.
In the third mode, as indicated by a thick broken line, information is acquired only from the WEB server without accessing the in-memory DB 14 in accordance with the analysis result of the request from the client 2, and is obtained from the client 2. It is a mode to transmit to. The third mode is a search operation for information on a WEB site that has been generally performed conventionally. This will be described in detail with reference to FIG.
In FIG. 10, the replica processing mode for the in-memory DB 14 performed between the DB server 18 and the search server 10 has been described with reference to FIG.

  First, the information search processing operation in the first mode will be described with reference to FIG. In the description of the processing operation, FIG. 7 and FIG. 8 may be referred to as needed for related explanation.

  The proxy processing function 12 of the DB search server loads the in-memory DB search list A shown in FIG. 7 into a memory (not shown) at the time of startup, and sets a request proxy URL, proxy SQL, and a search result template HTML. Set up. The in-memory DB search list A is a table that defines the contents of information registered in the in-memory DB 14. Normally, the WEB server 16 receives an HTTP request from the client 2 and issues a request to the DB server 18 based on the HTTP request to acquire the target information, but is defined in the in-memory DB search list A. For searching for specific information, the DB search server 10 executes the search instead of the information search by the DB server 18 without accessing the WEB server 16.

In the example shown in FIG. 7, the in-memory DB search list A is “http://WWW.WEB-AP.co.jp/cgi-bin/list1.cgi” as the URL of the WEB site acting on behalf of the HTTP request. Is described. The proxy SQL indicates that “customer name” and “service name” are searched from the “customer information” table under the condition of TEL = [$ TEL $]. Further, “List1_Result.html” is defined as a template HTML for search results.
Hereinafter, similarly, information of a replicated DB composed of a plurality of sets is defined.

  Returning to FIG. 3, the proxy processing function 12 waits for an HTTP request from the client 2 when the initial setting is completed. When an HTTP request is received from the client 2 (S32), the HTTP request analysis unit 122 analyzes the content of the HTTP request (S33). In this analysis, it is also checked whether the URI according to the HTTP request is included in the in-memory DB search list A. As a result of the analysis, if all the URIs of the HTTP request exist in the in-memory DB search list A, the in-memory DB 14 is accessed.

When accessing the in-memory DB 14, the in-memory DB search request processing unit 123 is activated (S361), and the in-memory DB 14 is searched (S362).
The information obtained as a result of this search is converted into HTML format data defined in the in-memory DB search list A by the execution of the response HTML generation unit 124 (S37). Further, the HTML format data is compressed by the content compression unit 125 (S38) and transmitted to the client 2 as an HTTP response (S39).

  When the client 2 receives the compressed content as a response to the HTTP request, the client 2 decompresses it, sends it to the browser 21, and displays it on the display of the PC 20.

  As described above, according to the first mode, it is not necessary to transmit an HTTP request to the WEB server 16 to acquire information, so that a response can be returned to the browser of the client 1 in a short time. Moreover, since the WEB server 16 and the DB server 18 are not accessed, those loads are reduced.

  Next, the information search operation in the second mode will be described with reference to FIG. 4 is different from FIG. 3 in that operations in steps S401 to S404 are added. Since the others are the same, the duplicate description is omitted.

  As a result of the analysis of the HTTP request (S33), if all of the requested plurality of URIs do not exist in the in-memory DB search list A, the HTTP request is transmitted as it is to the WEB server 16 (S401). The WEB server 16 requests the DB server 18 to acquire target information, for example, contents such as images and music, by processing an application program (WEB-AP) provided in advance (S402).

  Information acquired from the WEB server 16 is represented as a description in HTML format according to the definition of a response (extended tag) as shown in FIG. In the example of FIG. 8, it is indicated that data in a range surrounded by <XQUIX> to </ XQUIX>, that is, “customer name” and “service name” can be acquired from the DB replica unit 142 of the in-memory DB 14. Other data is acquired from the related DB by the execution of the DB server 18, and is described in a predetermined HTML format. Here, the reason for calling the “extended tag” is that the information included in the response from the WEB server 16 includes information for accessing the in-memory DB 14 in addition to the information described in the conventional HTML format. Because it is.

  When the proxy processing function 12 receives a response including an extension tag from the WEB server 16 (S403), the proxy processing function 12 analyzes the response (S404). As a result of the analysis, the in-memory DB search request processing unit 123 is activated for “customer name” and “service name” described as an extension tag in the response (S361). Then, the in-memory DB search unit 141 of the in-memory DB 14 is executed, and the “customer name” DB and “service name” DB registered in the DB replica unit 142 are searched to obtain target information (S362). ).

In this way, a response HTML is created for information on the target customer name and service name acquired from the in-memory DB (S37). Here, the response transmitted to the client 2 includes information acquired from the WEB server 16, but since the information acquired from the WEB server 16 has already been described in the HTML format, the extension tag Is sent to the client 2 as it is.
In the example of FIG. 4, the information obtained from the in-memory DB 14 and HTML-formatted and the HTML-formatted information obtained from the WEB server 16 are compressed by the content compression unit 125 (S38), and the HTTP response To the client 2 (S39).

  As described above, according to the information search processing in the second mode, an HTTP request in a case where information existing in the in-memory DB 14 (for example, information that takes time to search) and information not existing in the in-memory DB 14 are mixed. On the other hand, the overall time for information acquisition can be shortened.

Next, the information search operation in the third mode will be described with reference to FIG.
The third mode is an information acquisition operation from the WEB server 16 that has been generally performed conventionally. In this case, the HTTP request from the client is an access to a URL not described in the in-memory DB search list A.

  That is, as a result of analyzing the HTTP request from the client 2 (S33), if not all of the requested URIs exist in the in-memory DB search list A, the HTTP request is transmitted as it is to the WEB server 16 (S401). The WEB server 16 acquires target information from the DB server 18 by processing of an application program (WEB-AP) provided in advance (S402). Then, the response is transmitted to the proxy processing function 12 of the DB search server 10.

  The proxy processing function 12 receives a response from the WEB server 16 (S403). The subsequent operation is the same as described above, and the HTTP response analysis unit 126 of the proxy processing function 12 analyzes the received response. In this case, since it does not include a tag for searching the in-memory DB 14, the WEB server The response received from 16 is subjected to content compression processing (S38) and transmitted to the client 2 (S39).

Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made.
For example, the format of the list shown in FIGS. 7 to 9 is an example, and is not limited to these.

The figure which shows the structure of the WEB system in one Example of this invention. The figure which shows the module structure of DB search server 10 in the remote site of the WEB system by one Example. The figure which shows the process sequence of the information search operation | movement by the 1st mode in the WEB system by one Example. The figure which shows the process sequence of the information search operation | movement by the 2nd mode in the WEB system by one Example. The figure which shows the process sequence of the information search operation | movement by the 3rd mode in the WEB system by one Example. The figure which shows an example of the processing sequence of DB replica creation in the WEB system by one Example. The figure which shows the structural example of the in-memory DB search list A in the initial process of the proxy processing function in one Example. The figure which shows the example of a definition sample of the response tag of WEB-AP in one Example. The figure which shows the structural example of the replica condition list | wrist in the initial process of in-memory DB in one Example. The figure which shows the whole mode of processing operation of the information search in the WEB system by one Example.

Explanation of symbols

1: Remote site 2: Local site (client)
10: DB search server 12: Proxy processing function 14: In-memory 16: WEB server 18: DB server 20: PC
21: Browser

Claims (10)

An information search processing method in a WEB system for searching WEB page information on a search server based on an HTTP request from a client,
Storing specific information as a replica in the in-memory DB based on the conditions described in the replica condition list among the WEB page information obtained by executing the DB server;
The URI related to the WEB page information to be acquired included in the HTTP request from the client is analyzed, and when a specific URI is included in the HTTP request, a request is made to search the in-memory DB. Steps,
When the URI included in the HTTP request is a URI other than the specific URI, transmitting the HTTP request to the WEB server so as to obtain the WEB page information from the WEB server;
Searching for information stored in the in-memory DB for the specific URI according to the search request ;
Converting information acquired from the in-memory DB or the WEB server into data in HTML format;
Transmitting the converted HTML data to the client as an HTTP response ;
A search processing method comprising: acquiring the specific information from a DB server and rewriting and updating the contents of the in-memory DB at a predetermined time specified by the conditions described in the replica condition list . When the HTTP request is transmitted to the WEB server,
Analyzes whether the response from the WEB server includes an extension tag indicating information for accessing the in-memory DB in addition to the information described in the HTML format.
2. The search processing method according to claim 1 , wherein if the response includes an extension tag as a result of the analysis, the in-memory DB is searched by making a search request to the in-memory DB . The HTTP request sent from the client includes a URI specifying the WEB site,
A URI for designating a WEB site related to specific information stored in the in-memory DB, and an in-memory DB search list for defining the content of the specific information are stored in the memory.
It is checked whether or not the specific URI included in the HTTP request transmitted from the client is included in the in-memory DB search list. As a result of the check, the specific URI is detected in the in-memory DB search list. 3. The search processing method according to claim 1, further comprising: requesting to search the in-memory DB if it is defined in the list. The in-memory DB stores information of a type that takes time for the WEB server to acquire target WEB page information from the DB server in accordance with the conditions described in the replica condition list. The search processing method according to any one of items 1 to 3. According to the description condition of the replica condition list , the user information including any of ID, user name, and telephone number related to the user of the WEB system is acquired from the extraction source DB and stored in the in-memory DB. The search processing method according to claim 1, wherein: Search program for executing the method according to one of claims 1 to 5 in the search server. A WEB system having a search server that executes an application program on a WEB server based on an HTTP request from a client and obtains target WEB page information from a DB server ,
An in-memory DB that stores specific information as a replica based on the conditions described in the replica condition list among the WEB page information obtained by executing the DB server;
The URI related to the WEB page information to be acquired included in the HTTP request from the client is analyzed, and if the specific URI is included in the HTTP request, a request is made to search the in-memory DB. An HTTP request analysis unit that issues the HTTP request to the WEB server so that the web page information is acquired from the WEB server when the URI included in the HTTP request is a URI other than the specific URI;
An in-memory DB search request processing unit that issues a search request for searching the in-memory DB in accordance with the request from the HTTP request analysis unit;
In order to send the information acquired by searching the in-memory DB or the information acquired from the WEB server to the client, the response HTML generation unit edits the information into an HTML format, and is generated by the response HTML generation unit. Information in HTML format is sent to the client as an HTTP response,
The in-memory DB acquires the specific information from the DB server and rewrites and updates the contents of the in-memory DB at a predetermined time specified by the conditions described in the replica condition list. WEB system to do. The replica condition list includes a data extraction interval and extraction condition from the target DB, a description of the extraction source DB,
When the time specified by the extraction interval described in the replica condition list is reached, the in-memory DB acquires specific information from the extraction source DB according to the extraction condition, and rewrites the contents of the in-memory DB. The WEB system according to claim 7, wherein the WEB system is updated . The search server has a memory for storing an in-memory DB search list that defines the content of the specific information stored in the in-memory DB,
The HTTP request analysis unit analyzes whether the URI included in the received HTTP request is defined in the in-memory DB search list, and as a result of the analysis, the URI of the HTTP request is the in-memory DB search list. The in-memory DB search request processing unit issues a search request to the in-memory DB,
The WEB system according to claim 7 or 8, wherein the in-memory DB searches for information according to the search request . The HTTP request analysis unit, the in-memory DB search request processing unit, and the response HTML generation unit are provided in a proxy processing function realized by executing a program . The WEB system described in the section.

Images