JP2006106867A - Cache-provided searching device, cache optimization method applied to device, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To execute high-speed searching even if a use environment changes. <P>SOLUTION: A cache-provided searching device 10 comprises; an environment monitoring part 24 which monitors a use environment, in which its own device 10 is used, when a searching request is inputted from an input/output terminal 12, and determines whether the use environment thus monitored has changed; a cache log database 30 which stores the logs of the contents of a searching cache 32 in association with the use environment monitored by the environment monitoring part 24; and a cache-operating part 28 which obtains a log most closely relating to a changed environment from the cache log database 30 when the environment monitoring part 24 determines that the use environment has changed, and updates development data in the searching cache 32, based on the log thus obtained. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a search apparatus with a cache that is provided with a cache to improve the search speed, a cache optimization method applied to the search apparatus, and a program for causing a computer to realize its function.

Usually, a search device stores data to be searched in a database, and the corresponding data is searched from one or a plurality of databases based on a search request from a user. However, the database to be searched uses a secondary storage device such as a disk in order to store a large amount of data. Similar to a normal database management system, such a search device has a memory on which frequently searched data can be read faster than the secondary storage device in order to improve the search speed. Stored in the data cache. As a known example of such a cached search apparatus, there is, for example, Patent Document 1.
Japanese Patent Application No. 10-240766

  However, such a conventional cached search apparatus has the following problems.

  That is, in the conventional cached search device, since the index information of the most recently searched query is stored in the cache, immediately after the search device is started or immediately after the user starts the search, There is a problem that the cache cannot be effective because the query is not searched for.

  The present invention has been made in view of such circumstances. For example, when a usage environment such as a user or time information is recorded in association with a log of cache contents, the usage environment changes. A cache-equipped search device that can execute a high-speed search even when the usage environment changes by storing data that is expected to be searched in the cache in advance, and is applied to the search device. An object of the present invention is to provide a cache optimization method and program.

  In order to achieve the above object, the present invention takes the following measures.

  That is, in the cached search device according to the present invention, a part of the data stored in the storage means in advance is expanded in the cache on the memory, and when a search request is input from the outside, A cache optimization method that performs a high-speed search by using a cache and has the following functions is applied.

  This cache optimization method monitors the usage environment in which the device itself is used when a search request is input. Here, the usage environment includes, for example, a combination of users who make a search request and time information. Then, the log of the contents of the cache is stored in the log storage database in association with the monitored usage environment. Furthermore, it is determined whether there is a change in the usage environment being monitored. If it is determined that there has been a change in the usage environment, the log most relevant to the usage environment after the change is acquired from the log storage database. The expanded data in the cache is updated based on the acquired log. Here, when the change of the usage environment is determined, for example, when the combination of the searching users is changed or when a predetermined time is reached, it is determined that the usage environment has changed. Alternatively, the contents of the past log stored in the log storage database are analyzed in time series, and it is determined from this analysis result whether there is a change in the current use environment.

  Such a search apparatus with a cache can be configured by a program that realizes such a cache optimization method.

  According to the present invention, for example, immediately after startup, when a new user logs in, or when the usage environment changes, such as when a predetermined time comes, data that is expected to be retrieved is stored in the cache in advance. Can be stored. As described above, it is possible to realize a cached search apparatus capable of executing a high-speed search even when the usage environment changes, a cache optimization method, and a program applied thereto.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  FIG. 1 is a functional block diagram showing a configuration example of a cached search apparatus to which a cache optimization method according to an embodiment of the present invention is applied.

  That is, the cached search device 10 to which the cache optimization method according to the embodiment is applied includes a search server 20, an input / output terminal 12 and a database 14 connected thereto. Although there can be a plurality of input / output terminals 12 and databases 14, only one each is shown in FIG. 1 for the sake of simplicity of explanation. The input / output terminal 12 sends a search request to the search server 20 or displays a search result sent from the search server 20, and includes input tools such as a keyboard and a mouse, and data input by the input tool. And a display for displaying search results from the search server 20. Such an input / output terminal 12 can be constituted by a personal computer, for example. Further, it may be remotely located from the search server 20 via a communication network such as the Internet.

  The search server 20 further includes a cache analysis unit 22, an environment monitoring unit 24, a search server information unit 26, a cache operation unit 28, a cache log database 30, a general search cache 32, and a search execution unit 34. It has.

  The cache analysis unit 22 designates a time interval for recording the cache log, and sends the designated time interval to the environment monitoring unit 24. Note that the cache analysis unit 22 may be omitted, the time interval for recording the cache log may be specified from the input / output terminal 12, and the specified time interval may be sent to the environment monitoring unit 24.

  In addition, the cache analysis unit 22 analyzes the past cache log stored in the cache log database 30 when the search server 20 is started up, and the optimal timing for replacing the cache expansion data recorded in the search cache 32. Is automatically calculated. This calculation method will be described with reference to the flowchart shown in FIG.

  That is, the cache analysis unit 22 compares all the cache logs stored in the cache log database 30 at the time when the logs are recorded, and groups them as a group (S1).

Next, all the cache logs in each group grouped in step S1 are expanded into a vector space (S2). In the vector space, all registered words are taken as dimensions, and 1 is entered if the word exists, and 0 is entered if the word does not exist. For example, as shown in FIG. 3, the cache logs 1 and 2 are grouped in step S 1, and the words “Japan” and “Economy” are recorded in the cache log 1. When the words “Japan” and “baseball” are recorded, all the registered words are “Japan”, “economy”, and “baseball”. Therefore, the vector space is defined by a three-dimensional space (Japan, economy, baseball). For such a vector space, the result of vector space expansion of cache log 1 in which “Japan” and “economy” are recorded is
(Japan, economy, baseball) = (1, 1, 0)
The vector space expansion result of cash log 2 where “Japan” and “baseball” are recorded is
(Japan, economy, baseball) = (1, 0, 1).

Next, for each group, the sum of all vectors in the group is taken. The sum of these vectors is called a representative vector (S3). In the case of the cache log database 30 in which the cache log 1 and the cache log 2 are recorded as shown in FIG.
(Japan, economy, baseball) = (1,1,0) + (1,0,1) = (2,1,1).

  Next, in each representative vector obtained in step S3, a comparison value having a magnitude as shown in the following equation (3) with respect to the previous time and the next time, shown in equation (4) A comparison value in each direction is calculated. In equations (3) and (4), vector A is a representative vector of the previous time, and vector B is a representative vector of the next time. When either one of the comparison values obtained according to the formula (3) or the formula (4) exceeds a predetermined threshold, the time at the boundary between the two groups is used as the cache replacement timing, and the environment is changed. The data is output to the monitoring unit 24 (S4). By changing such processing so as to be grouped for each user, the timing of cache replacement is automatically calculated.

  The search server information unit 26 provides time information to the environment monitoring unit 24.

  The environment monitoring unit 24 grasps the current time by acquiring time information from the search server information unit 26. Each time the designated time interval from the cache analysis unit 22 or the input / output terminal 12 is reached, an instruction signal is sent to the cache operation unit 28 to record the log of the contents in the search cache 32 in the cache log database 30. send. In response to this, the cache operation unit 28 records a log of the contents in the search cache 32 in the cache database 30.

  The environment monitoring unit 24 also monitors the usage environment in which the search server 20 is used when a search request is input from the input / output terminal 12. The usage environment includes a user who operates the input / output terminal 12 and time information. Then, it is determined whether or not the usage environment being monitored has changed. For example, when the search server 20 is activated or when a predetermined time is reached, it is detected and it is determined that the use environment has changed. Further, the contents of the past log stored in the cache log database 30 are analyzed in time series, and it is determined from the analysis result that the current usage environment has changed. When it is determined that the usage environment has changed in this way, the first cache replacement signal is output to the cache operation unit 28 every time the cache replacement timing calculated by the cache analysis unit 22 is reached.

  Further, when the user operating the input / output terminal 12 is changed to another user, it is determined that the use environment has changed. In this case, a second cache replacement signal is output to the cache operation unit 28 every time the cache replacement timing calculated by the cache analysis unit 22 is reached.

  When the cache replacement signal is output from the environment monitoring unit 24, the cache operation unit 28 acquires the cache log most relevant to the current usage environment from the cache log database 30. For example, if the first cache replacement signal is sent, the most relevant cache log is acquired from the cache log database 30 based on the time. Specifically, the cache log having the highest evaluation score from the following evaluation formula (1) is acquired from the cache logs dated before yesterday. However, in the evaluation formula (1), the time is “hour: minute: second” excluding the year and the day of the recording time. This time can be expressed as YY: MM: DD: hh: mm: ss (YY is the last two digits of the year, MM is the month, DD is the day, hh is the hour, mm is the minute, and ss is the second). As an example, the operation is performed for each element separated by: If the value of the element becomes negative, it can be borrowed from a higher element. The time difference is expressed in units of 10 minutes, and the date difference is expressed in days.

Evaluation point = (current time + 10 minutes) − (log time)... Evaluation formula (1)
The meaning of this evaluation formula (1) is to select the latest cache log at the nearest time after 10 minutes.

  Further, if the second cache replacement signal is output, the cache operation unit 28 selects the cache log having the lowest evaluation score from the following evaluation formula (2), so that Among them, the cache log having the most similar user configuration is acquired from the cache log database 30. Note that the priority of the user used in the evaluation formula (2) (the higher the value, the higher) is determined in advance.

Evaluation point = (User composition point) * ((Current date) − (Date of log))... Evaluation formula (2)
Here, the user composing point is obtained from the sum of the absolute values of each element of the user vector between the current and the cache log. Below, a user vector and a user component point are demonstrated concretely.

That is, when calculating an evaluation score using the evaluation formula (2), first, a vector having a dimension corresponding to the total number of users is considered, and a user vector in the current situation is created. In the creation method, each dimension of the vector is assigned to one user, and the priority is assigned if the user is logged in, and 0 is assigned if the user is not logged in. For example, for four users W, X, Y, and Z, the priority of user W is 20, the priority of user X is 10, the priority of user Y is 10, and the priority of user Z is 5. In this case, the user vector when only the user W and the user Y are logged in is
User vector (W, X, Y, Z) = (20, 0, 10, 0).
Moreover, the user vector in the state where only the user X, the user Y, and the user Z are logged in is
User vector (W, X, Y, Z) = (0, 10, 10, 5).

Next, a specific evaluation point calculation example using the evaluation formula (2) will be described. As a first assumption case, all of the users W, X, Y, and Z are currently logged in, and only one user Z with a low priority is not logged in in the user situation of a certain log α one day ago. Think about the case. In this case, the current user vector is (W, X, Y, Z) = (20, 10, 10, 5), and the user vector in the log α is (W, X, Y, Z) = (20 , 10, 10, 0), the difference between each element of the user vector between the current and the cache log is
(ΔW, ΔX, ΔY, ΔZ) = (0, 0, 0, 5). Therefore, the sum of absolute values of each element is 0 + 0 + 0 + 5 = 5. This is the user composition point. Since ((current date) − (log date)) in the evaluation formula (2) is 1 (day), the evaluation score in this case is 5 × 1 = 5 points.

  On the other hand, as a second assumed case, all of the users W, X, Y, and Z are currently logged in, and in the user situation of a certain log β two days ago, only the user W with a high priority logs in. Think if not. In this case, the current user vector is (W, X, Y, Z) = (20, 10, 10, 5), and the user vector in the log β is (W, X, Y, Z) = (0 , 10, 10, 5), the difference between each element of the user vector between the current and the cache log is (ΔW, ΔX, ΔY, ΔZ) = (20, 0, 0, 0). Therefore, the sum of absolute values of each element is 20 + 0 + 0 + 0 = 20. This is the user composition point. Since ((current date) − (log date)) in the evaluation formula (2) is 2 (day), the evaluation score in this case is 20 × 2 = 40 points.

  When the log α and the log β calculated as described above are compared, the log α has a lower evaluation score, so the user configuration of the log α is determined to be the most similar cache log.

  When the cache log is acquired in this way, the cache operation unit 28 further deletes 50% of the words currently cached in the search cache 32 from the lower priority words. Of the words recorded in the cache log acquired from the cache log database 30, 50% from the word with the highest priority is inserted as 50% from the word with the lowest priority in the search cache 32. Among the words to be inserted, words that are already included in 50% from the highest priority cached in the search cache 32 are left as they are currently cached in the search cache 32. Of the words recorded in the cache log, among the words that have not been inserted into the search cache 32, the word having the highest priority is moved up and inserted into the search cache 32. In this way, the data in the search cache 32 is updated based on the acquired contents of the cache log.

  This will be described with reference to FIG. FIG. 4A is a diagram showing words currently cached in the search cache 32 and their priorities. On the other hand, FIG. 4B is a diagram showing the contents of the cache log acquired from the cache log database 30. In this case, the cache operation unit 28 deletes 50% of the words cached in the search cache 32 as shown in FIG. That is, the search cache 32 caches four words of “Japan”, “Prime Minister”, “Tokyo”, and “Prime Minister” in descending order of priority, so that the words are 50% of low priority. “Tokyo” and “Prime Minister” are deleted, and “Japan” and “Prime Minister” remain as high-priority 50% words.

  On the other hand, as shown in FIG. 4B, the words “Japan”, “baseball”, “MLB”, and “home run” are recorded in the cache log acquired from the cache log database 30 in descending order of priority. Therefore, “Japan” and “baseball”, which are words with a high priority of 50%, are inserted into the lower priority 50% of the search cache 32. However, since the word “Japan” already remains in the top 50% of the priorities in the search cache 32, the word “MLB” is carried forward.

  As a result, as shown in FIG. 4C, among the words recorded in the search cache 32, “Japan” and “Prime Minister”, which are high-priority words of 50%, remain as they are. The words “Tokyo” and “Prime Minister”, which are 50% lower words, are deleted from the search cache 32. Instead of the words included in the cache log acquired from the cache log database 30, the top 50% of priorities excluding the words already recorded in the search cache 32 (“Japan”, “Prime Minister”) The words “baseball” and “MLB” are inserted in the lower priority 50% of the search cache 32.

  When an instruction signal is sent from the environment monitoring unit 24, the cache operation unit 28 also includes a recording time recorded in the cache log database 30 and a user name currently using the input / output terminal 12 as shown in FIG. A cache log including the environment, the words cached in the search cache 32 and the search frequency order is created and written to the cache log database 30.

  The search execution unit 34 executes a search from the search cache 32 or the database 14 as follows based on the search request sent from the input / output terminal 12 by the user. That is, when a search request is sent from the input / output terminal 12, an attempt is made to search the search cache 32 for the word requested to be searched. For example, as shown in FIG. 6, the search cache 32 records words according to the priority order sorted in descending order of search frequency. In addition, as index information, a document name in which a word is described and a description location are recorded.

  If the search requested word is stored in the search cache 32, the index corresponding to the word is read out, a list of search results is created, and the result is returned to the input / output terminal 12. Thereafter, as shown in FIG. 7 as an example, in the search cache 32, an update process for lowering the priority of all words higher than the searched word one by one and setting the priority of the searched word to first is performed. Execute.

  FIG. 7 shows an example of update processing of the search cache 32 when searching for a search request “Tokyo” and document # 8 is searched from the search cache 32. That is, before the search, as shown in FIG. 7A, the word “Tokyo” was in the third priority, but after the search, as shown in FIG. The priority of “Tokyo” has been raised to first place, and the priority of other words has been lowered by one.

  On the other hand, if the search requested word is not stored in the search cache 32, the index corresponding to the word is read from the database 14, a list of search results is created, and the result is input to the input / output terminal 12. Return to. Thereafter, as shown in FIG. 8 as an example, an update process for deleting the index of the word with the lowest priority in the search cache 32 and setting the priority of the index of the word searched from the database 14 to first place is performed. Execute.

  FIG. 8 shows an example of update processing of the search cache 32 when a search is performed in response to the search request “Bush” and the document # 10 is searched from the database 14. That is, before the search, as shown in FIG. 8A, the word “bush” was not recorded in the search cache 32, but after the search, As shown, the priority of “Bush” has been raised to the first place, and instead the word “Tokyo” with the lowest priority before the search has been deleted.

  In the above description, for the sake of simplicity, the case where the search request is made with a single word has been described as an example. However, the search request is not limited to a single word, and a plurality of words, Or it is applicable similarly even if it is a sentence. Also, when the data in the database 14 is updated, the contents of the search cache 32 may change, but this is omitted for the sake of brevity.

  The cached search device 10 having such a configuration reads a program stored in a storage medium such as a magnetic disk or a program downloaded via a network such as the Internet, and a computer whose operation is controlled by the program. It is realized by.

  Here, the storage medium can store programs such as magnetic disk, floppy disk, hard disk, optical disk (CD-ROM, DVD, etc.), magneto-optical disk (MO, etc.), semiconductor memory, etc., and can be read by a computer. As long as it is a possible storage medium, the storage format may be any form.

  Further, an OS (operating system) running on the computer based on an instruction of a program installed in the computer from the storage medium, MW (middleware) such as database management software, network software, and the like realize the embodiment. A part of each process may be executed.

  Furthermore, the storage medium is not limited to a medium independent of the computer, but also includes a storage medium in which a program transmitted via a LAN, the Internet, or the like is downloaded and stored or temporarily stored.

  Further, the number of storage media is not limited to one, and the case where the processing in the embodiment is executed from a plurality of media is also included in the storage media here, and the media configuration may be any configuration.

  The computer referred to here executes each processing in the embodiment based on a program stored in a storage medium, and a single device such as a personal computer or a plurality of devices are connected to a network. Any configuration such as a system may be used. In addition, the computer is not limited to a personal computer, but includes an arithmetic processing device, a microcomputer, and the like included in an information processing device, and is a generic term for devices and devices that can realize the functions of the present invention by a program.

  Next, an operation of the cached search device 10 to which the cache optimization method according to the embodiment configured as described above is applied will be described.

  That is, in the cached search device 10 to which the cache optimization method according to the embodiment is applied, the cache analysis unit 22 specifies a time interval for recording the cache log, and the specified time interval is sent to the environment monitoring unit 24. It is done. Note that the cache analysis unit 22 may be omitted, the time interval for recording the cache log may be specified from the input / output terminal 12, and the specified time interval may be sent to the environment monitoring unit 24.

  The cache analysis unit 22 analyzes the past cache log stored in the cache log database 30 when the search server 20 is started, and the cache expansion data recorded in the search cache 32 as shown in the flowchart of FIG. The optimum timing for replacing is automatically calculated.

  Further, the search server information unit 26 provides time information to the environment monitoring unit 24.

  Then, the environment monitoring unit 24 grasps the current time from the time information provided from the search server information unit 26. Each time the specified time interval from the cache analysis unit 22 or the input / output terminal 12 is reached, an instruction signal is sent to the cache operation unit 28 to record the log of the contents in the search cache 32 in the cache log database 30. Sent. In response to this, the cache operation unit 28 records a log of the contents in the search cache 32 in the cache log database 30.

  In addition, when a search request is input from the input / output terminal 12, the environment monitoring unit 24 monitors the use environment in which the search server 20 is used, and determines whether or not the use environment has changed. For example, when the search server 20 is activated or when a predetermined time is reached, it is detected and it is determined that the use environment has changed. Further, the contents of past logs stored in the cache log database 30 are analyzed in time series, and it is determined from the analysis result that there is a change in the current use environment. When it is determined that the usage environment has changed in this way, the first cache replacement signal is sent to the cache operation unit 28 every time the cache replacement timing calculated by the cache analysis unit 22 is reached. .

  Also, when the user operating the input / output terminal 12 changes to another user, it is determined that the usage environment has changed. In this case, the second cache replacement signal is sent to the cache operation unit 28 every time the cache replacement timing calculated by the cache analysis unit 22 is reached.

  In the cache operation unit 28, when a cache replacement signal is sent from the environment monitoring unit 24, the cache log most relevant to the current use environment is acquired from the cache log database 30 according to the flowchart shown in FIG. Is done. For example, when the search server 20 is activated (S11), the first cache replacement signal is sent from the environment monitoring unit 24, and is most relevant according to the evaluation formula (1) described above based on the time. A cache log is selected from the cache log database 30 (S12). Then, the cache data in the search cache 32 is replaced by the cache operation unit 28 in accordance with the selected cache log (S13), and then the environment monitoring by the environment monitoring unit 24 is continued (S14).

  Further, in this state, when the designated time for selecting the cache log is reached (S15: Yes), the processing from step S12 is similarly performed, and thereafter, the environment monitoring by the environment monitoring unit 24 is continued (S14). ). On the other hand, even if it is not the designated time for selecting the cache log (S15: No), when the new user logs in to the search server 20 (S16: Yes), the environment monitoring unit 24 2 cache replacement signals are sent, and the cache log with the lowest user score is selected from the recent cache logs by selecting the cache log with the lowest evaluation score according to the above-described evaluation formula (2). It is selected from the cache log database 30 (S17). Then, the cache data in the search cache 32 is exchanged according to the selected cache log (S13), and then the environment monitoring by the environment monitoring unit 24 is continued (S14). On the other hand, in step S16, when a new user does not log in (S16: No), the environment monitoring by the environment monitoring unit 24 is continued (S14).

  Each time the instruction signal is sent from the environment monitoring unit 24, the cache operation unit 28, as shown in FIG. 5, from the recording time recorded in the cache log database 30 and the name of the user currently using the input / output terminal 12. A cache log including the usage environment, the words cached in the search cache 32 and the search frequency order is created and written to the cache log database 30.

  On the other hand, the search execution unit 34 executes a search from the search cache 32 or the database 14 as shown in the flowchart of FIG. 10 based on the search request sent from the input / output terminal 12 by the user. That is, when a search request is input from the input / output terminal 12 (S21), a search for the requested word from the search cache 32 is attempted. In the search cache 32, for example, as shown in FIG. 6, words are recorded in the order of priority sorted in descending order of search frequency. In addition, as index information, a document name in which a word is described and a description location are recorded.

  If the word requested for search in step S21 is stored in the search cache 32 (S22: Yes), an index corresponding to the word is acquired from the search cache 32, and a list of search results is created. (S23). Thereafter, as shown in FIG. 7 as an example, in the search cache 32, an update process for lowering the priority of all the words higher than the searched word one by one and setting the priority of the searched word to the first is performed. As a result, the search cache 32 is reconstructed (S24). A list of search results created in step S23 is displayed from the input / output terminal 12 (S25).

  On the other hand, if the word requested to be searched in step S21 is not stored in the search cache 32 (S23: No), the index corresponding to the word is read from the database 14, and a list of search results is displayed. It is created (S26). A list of the created search results is displayed from the input / output terminal 12 (S27).

  As described above, in the cached search device 10 to which the cache optimization method according to the embodiment is applied, for example, when the search server 20 is started, when a new user logs in, When the usage environment changes due to the arrival of the designated time, the environment monitoring unit 24 can determine it.

  If it is determined that the usage environment has changed, an optimal cache log is acquired from the cache log database 30 according to the change, and the expanded data in the search cache 32 is updated based on the acquired cache log. be able to.

  As a result, when the usage environment changes, the expanded data of the search cache 32 can be updated to contents suitable for the changed usage environment, so even if the usage environment changes, high-speed search is possible. Can be executed.

  The best mode for carrying out the present invention has been described above with reference to the accompanying drawings, but the present invention is not limited to such a configuration. Within the scope of the invented technical idea of the scope of claims, a person skilled in the art can conceive of various changes and modifications. The technical scope of the present invention is also applicable to these changes and modifications. It is understood that it belongs to.

The functional block diagram which shows the structural example of the search apparatus with a cache to which the cache optimization method which concerns on embodiment of this invention is applied. The flowchart which shows an example of the calculation method of the replacement | exchange optimal timing of cache expansion | deployment data. The figure which shows an example of the word currently recorded for every cash log group. The figure for demonstrating an example of the update method of the cache made | formed by a cache operation part. The figure which shows the structural example of a cache log. The figure which shows the data structural example of a search cache. The figure for demonstrating the update process of the search cache performed in a search process (when the search requested word exists in a search cache). The figure for demonstrating the update process of the search cache performed in a search process (when the search requested word does not exist in a search cache). The flowchart which shows an example of the flow of a process in the case of replacing the cache data in a search cache. The flowchart which shows an example of the flow of a search process.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Search apparatus with cache, 12 ... Input / output terminal, 14 ... Database, 20 ... Search server, 22 ... Cache analysis part, 24 ... Environment monitoring part, 26 ... Search server information part, 28 ... Cache operation part, 30 ... Cache Log database, 32 ... search cache, 34 ... search execution unit

Claims (9)

A part of the data stored in the storage means is expanded in a cache on the memory, and when a search request is input from the outside, the cache is used to perform a high-speed search. In the search device with cache,
Environment monitoring means for monitoring a use environment in which the device itself is used when the search request is input;
Log storage means for storing the log of the contents of the cache in association with the usage environment monitored by the environment monitoring means;
Environment change determination means for determining whether or not the use environment monitored by the environment monitoring means has changed;
When it is determined by the environment change determination means that the use environment has changed, a log acquisition means for acquiring a log most relevant to the use environment after the change from the log storage means;
A cached search device comprising: a cache update unit that updates the expanded data in the cache based on the log acquired by the log acquisition unit. The retrieval apparatus with a cache according to claim 1,
The use environment is a search device with cache including a user who makes the search request and time information. The retrieval apparatus with a cache according to claim 2,
The search apparatus with cache, wherein the environment change determination means determines that the use environment has changed when the user changes to another user or when a predetermined time comes. The retrieval apparatus with a cache according to claim 1,
The environment change determination unit analyzes the contents of past logs stored in the log storage unit in a time series, and determines whether there is a change in the current use environment from the analysis result. Search device with cache. A part of the data stored in the storage means is expanded in a cache on the memory, and when a search request is input from the outside, the cache is used to perform a high-speed search. A cache optimization method applied to a cached search device,
Monitoring the usage environment in which the cached search device is used when the search request is input;
Storing a log of the contents of the cache in a log storage database in association with the monitored usage environment;
Determining whether there has been a change in the monitored use environment;
If it is determined that there has been a change in the usage environment, obtaining a log most relevant to the usage environment after the change from the log storage database;
A cache optimization method comprising: updating expanded data in the cache based on the acquired log. The cache optimization method according to claim 5, wherein
The usage environment is a cache optimization method including a user who makes the search request and time information. The cache optimization method according to claim 6, wherein
A cache optimization method in which it is determined that the usage environment has changed when the user changes to another user or when a predetermined time comes. The cache optimization method according to claim 5, wherein
A cache optimization method in which the contents of past logs stored in the log storage database are analyzed in time series, and it is determined from this analysis result whether or not the current usage environment has changed. A part of the data stored in the storage means is expanded in a cache on the memory, and when a search request is input from the outside, the cache is used to perform a high-speed search. A program that is applied to a cached search device,
A function for monitoring a use environment in which the cached search device is used when the search request is input;
A function of storing a log of the contents of the cache in a log storage database in association with the monitored usage environment;
A function for determining whether or not there is a change in the monitored use environment;
If it is determined that there has been a change in the usage environment, a function for obtaining a log most relevant to the usage environment after the change from the log storage database;
A program for causing a computer to realize a function of updating the expansion data in the cache based on the acquired log.

Images