WO2019163610A1 - Information processing system and information processing method - Google Patents

Abstract

An information processing system is characterized by having: a storage means that stores a table in which a plurality of records are stored, first array data in which prescribed item values included in each of the plurality of records are stored, second array data in which a record number identifying the records is stored, and third array data associating the first array data and the second array data; and an acquisition means that, in response to a designation of an item value or a record number, uses the first array data, the second array data, and the third array data to acquire one or more records corresponding to the designated item value or record number among the plurality of records stored in the table.

Description

Information processing system and information processing method

The present invention relates to an information processing system and an information processing method.

2. Description of the Related Art Conventionally, a technique for searching data stored in a database using an index such as B-Tree or hash and displaying the search result is known (for example, Patent Documents 1 and 2).

JP 2007-25735 A JP-A-6-335165

By the way, in recent years, with the development of various sensor devices and observation devices, a large amount of data representing sensing results and observation results can be obtained. For this reason, it is required to search desired data from these large amounts of data (so-called big data) and display the search results.

However, big data is difficult to transfer and aggregate because of its data size. In addition, a search server for searching for desired data from big data generally needs to have high-performance hardware resources, and the installation cost is high. Furthermore, when a search server equipped with high-performance hardware resources is installed, its maintenance cost (for example, maintenance cost, power consumption, etc.) is also expensive.

For this reason, the big data provider wants to use a simple file server that does not include expensive hardware and software related to search, and present big data as a file. In this case, the maintenance is easy and the cost is low.

Also, the Big Data usage side is a general PC (personal computer), etc., and we want to be able to search by combining the Big Data provider's files freely and display the search results.

The present invention has been made in view of the above points, and enables a user of big data to freely combine a plurality of big data files distributed on a network such as the Internet and easily search and display desired data. To.

In order to solve the above problems, an embodiment of the present invention includes a table storing a plurality of records, first array data storing predetermined item values included in each of the plurality of records, Storage means for storing second array data in which a record number for identifying a record is stored, and third array data that associates the first array data with the second array data, and the item value or In response to the designation of the record number, the first array data, the second array data, and the third array data are used to specify the specified record among the plurality of records stored in the table. Obtaining means for obtaining one or more records corresponding to the item value or the record number.

Big data users can freely combine multiple big data files distributed over a network such as the Internet to easily search and display desired data.

It is a figure which shows an example of the whole structure of the data management system in embodiment of this invention. It is a figure which shows an example of the hardware constitutions of a computer. It is a figure which shows an example of a table. It is a figure which shows an example of a record order view. It is a figure which shows an example of a value order view. It is a figure which shows an example of a record order UNION view. It is a figure which shows an example of a value order UNION view. FIG. 3 is a diagram (part 1) illustrating an example of an internal data structure; It is FIG. (2) which shows an example of an internal data structure. It is a figure for demonstrating an example of the production method of SVL, IND, and INV. 10 is a diagram for explaining an application example 1. FIG. FIG. 10 is a diagram for explaining an application example 2;

Hereinafter, embodiments of the present invention (hereinafter also referred to as “present embodiments”) will be described. Hereinafter, the data management system 1 that displays and retrieves data stored in the database will be described.

<Overall configuration>
First, the overall configuration of the data management system 1 according to the present embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of the overall configuration of a data management system 1 according to an embodiment of the present invention.

As shown in FIG. 1, the data management system 1 according to the present embodiment includes one or more terminal devices 10 and one or more data management devices 20. Further, the terminal device 10 and the data management device 20 are communicably connected via a network N such as the Internet, a LAN (Local Area Network), or a telephone line network.

The terminal device 10 is an information processing device (computer) such as a PC that is operated by a user who uses big data. The user can display and search data managed by the data management device 20 by operating the terminal device 10.

The terminal device 10 includes a display processing unit 100 for displaying and retrieving data managed by the data management device 20. The display processing unit 100 is realized by, for example, a process in which one or more programs installed in the terminal device 10 are executed by a CPU (Central Processing Unit).

The display processing unit 100 creates and displays a view of data managed by the data management apparatus 20 in accordance with a user operation. Further, the display processing unit 100 searches for desired data from a certain view and displays the search result. A view is display information representing data arbitrarily selected from one or a plurality of tables (or other views). A table (or table) is information composed of one or more data. In addition, data search means that a certain item of each data constituting a certain view is designated from the view by specifying a range of values of the item (for example, =, <,>, between, etc.). Extracting data. Therefore, the result of data retrieval from a certain view is also a view. For this reason, a record order view, a value order view, a record order UNION view, and a value order UNION view, which will be described later, can be said to be views created by all the records constituting one or more tables 1000, while data The search result of the search can be said to be a view in which some records of these views are displayed.

The terminal device 10 is not limited to a PC, and may be various information processing devices such as a smartphone and a tablet terminal.

The data management device 20 is an information processing device (computer) such as a database server on which the big data provider manages various data. The data management device 20 includes a data processing unit 200 that manages various data, and a database 300 that stores various data. The data processing unit 200 is realized, for example, by a process that the CPU executes by one or more programs installed in the data management device 20. The database 300 can be realized by, for example, an auxiliary storage device.

The data processing unit 200 transmits the corresponding data among the data stored in the database 300 to the terminal device 10 in response to a request from the terminal device 10. Thereby, in the terminal device 10, a view is created and displayed, or a search result is displayed.

Here, the database 300 includes a table 1000 composed of one or more data, an SVL (Sorted Value List) 1100, an IND (Indirect Record Index) 1200, and an INV (Inverted Record Index) used for view creation and data retrieval. ) 1300 is stored. SVL 1100, IND 1200, and INV 1300 are internal data created with respect to a certain item of each record constituting table 1000. Details of these data will be described later.

The views displayed by the display processing unit 100 of the terminal device 10 according to the present embodiment include a record order view, a value order view, a record order UNION view, and a value order UNION view. The record order view is a view that displays each data constituting the table 1000 (this data is also referred to as “record”) in order of record number. The value order view is a view that displays the records constituting the table 1000 in the order of values of a specified item (this item is also referred to as “column”) of these records. The record order UNION view is a view that displays the records constituting the plurality of tables 1000 in the order of records. The value order UNION view is a view that displays each record constituting each of the plurality of tables 1000 in the order of values of a specified item of these records. Details of these views will be described later. Note that creating one view from a plurality of tables 1000 (or views) is also referred to as “UNION” or “virtual UNION”.

At this time, the record order UNION view and the value order UNION view may be created from the records constituting the table 1000 managed by the same data management apparatus 20, or are managed by different data management apparatuses 20, respectively. It may be created from records constituting the table 1000. Hereinafter, when different data management devices 20 are distinguished from each other, they are represented as a data management device 20-1, a data management device 20-2, and the like. Similarly, when different tables 1000 are distinguished, they are represented as table 1000-1, table 1000-2, and the like. The same applies when differentiating each different SVL 1100, each different IND 1200, and each different INV 1300.

The data management system 1 according to the present embodiment uses the SVL 1100, IND 1200, and INV 1300 when creating and displaying various views, so that even if the number of records constituting the table 1000 is enormous, Various views can be created and displayed at high speed. In addition, the data management system 1 according to the present embodiment uses the SVL 1100, the IND 1200, and the INV 1300 when performing a data search, so that the search can be performed at a high speed and the search result can be obtained with a small data capacity. Will be able to hold.

In the example shown in FIG. 1, the terminal device 10 and the data management device 20 are shown as different devices. However, the present embodiment is similarly applied even when these devices are integrally configured. Can be applied. Further, the data management device 20 may be realized by, for example, NAS (Network Attached Storage) or the like.

<Hardware configuration>
Next, the hardware configuration of one or more computers that implement the terminal device 10 and the data management device 20 according to the present embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating an example of a hardware configuration of the computer 500.

The computer 500 shown in FIG. 2 includes an input device 501, a display device 502, an external I / F 503, a RAM (Random Access Memory) 504, a ROM (Read Only Memory) 505, a CPU 506, and a communication I / F 507. And an auxiliary storage device 508. Each of these hardware is connected via a bus B so as to be able to communicate.

The input device 501 is, for example, a keyboard, a mouse, a touch panel, or the like, and is used for a user to input various operations. The display device 502 is a display, for example, and displays various processing results (for example, various views). Note that the data management device 20 may not include at least one of the input device 501 and the display device 502.

External I / F 503 is an interface with an external device. The external device includes a recording medium 503a and the like. The terminal device 10 and the data management device 20 can read and write the recording medium 503a and the like via the external I / F 503.

Examples of the recording medium 503a include a flexible disk, a CD (Compact Disc), a DVD (Digital Versatile Disk), an SD memory card (Secure Digital memory card), a USB (Universal Serial Bus) memory card, and the like.

The communication I / F 507 is an interface for connecting the computer 500 to the network N. The terminal device 10 and the data management device 20 can perform data communication with other devices via the communication I / F 507.

The RAM 504 is a volatile semiconductor memory that temporarily stores programs and data. The ROM 505 is a nonvolatile semiconductor memory that can retain programs and data even when the power is turned off. The ROM 505 stores, for example, OS (Operating System) settings, network settings, and the like.

The CPU 506 is an arithmetic device that reads out programs and data from the ROM 505, the auxiliary storage device 508, and the like onto the RAM 504 and executes processing.

The auxiliary storage device 508 is, for example, a hard disk drive (HDD) or a solid state drive (SSD), and is a non-volatile storage device that stores programs and data. Examples of the programs and data stored in the auxiliary storage device 508 include an OS and application software that implements various functions on the OS. Note that one or more programs for realizing the display processing unit 100 are stored in the auxiliary storage device 508 of the computer 500 that realizes the terminal device 10. Similarly, the auxiliary storage device 508 of the computer 500 that implements the data management device 20 stores one or more programs that implement the data processing unit 200.

The terminal device 10 and the data management device 20 according to the present embodiment can implement various processes according to the present embodiment by the hardware configuration of the computer 500 illustrated in FIG.

<Table>
Here, in the present embodiment, a case where each view is displayed using the table 1000-1 shown in FIG. 3A and the table 1000-2 shown in FIG. 3B will be described. Hereinafter, it is assumed that the table number of the table 1000-1 shown in FIG. 3A is “0” and the table number of the table 1000-2 shown in FIG. 3B is “1”. The table 1000 is an electronic file or data composed of one or more data (records). The table number is identification information for identifying at least the table 1000 used for creating and displaying a view.

3A is stored in, for example, the database 300 of the data management apparatus 20-1, while the table 1000-2 illustrated in FIG. 3B is, for example, the data management apparatus. It is assumed that it is stored in the database 300 of 20-2. However, the table 1000-1 shown in FIG. 3A and the table 1000-2 shown in FIG. 3B may be stored in the database 300 of the same data management apparatus 20.

The table 1000-1 shown in FIG. 3A includes four records, and each record includes an item “Name” and an item “Age”. Similarly, the table 1000-2 illustrated in FIG. 3B includes four records, and each record includes an item “Name” and an item “Age”. The position is a number indicating a relative position from the first record in each table 1000 (for example, an address (address) of a storage area in which each record is stored).

Note that each of the table 1000-1 and the table 1000-2 shown in FIGS. 3A and 3B is composed of four records, but this is an example, and each table 1000-1 The number of records that make up is arbitrary. When handling big data, the number of records constituting each table 1000 is generally tens of thousands to hundreds of millions or more. Each record constituting the table 1000-1 and the table 1000-2 shown in FIGS. 3A and 3B has two items, but this is an example. The number of items included in is also arbitrary.

<Each view>
Next, each view in the case where the table 1000-1 and the table 1000-2 shown in FIGS. 3A and 3B are used will be described.

≪Record order view≫
FIG. 4A shows a record order view VW1-1 of the table 1000-1 shown in FIG. FIG. 4B is a record order view VW1-2 of the table 1000-2 shown in FIG.

As described above, the record order view is a view that displays the records constituting the table 1000 in order. However, the record order view may display the records constituting the table 1000 in reverse order, for example.

It should be noted that each record displayed in the record order view VW1 shown in FIG. 4 includes all items of each record constituting the table 1000, but is not limited thereto. The items of each record displayed in the record order view may be some items of each record constituting the table 1000. The same applies to other views such as a value order view, a record order UNION view, and a value order UNION view.

≪Value order view≫
FIG. 5A shows a value order view VW2-1 regarding the item “Name” of the table 1000-1 shown in FIG. FIG. 5B is a value order view VW2-2 regarding the item “Name” of the table 1000-2 shown in FIG.

As shown in FIGS. 5A and 5B, the item “RecNo” is added to each record of the value order views VW2-1 and VW2-2. The value of the item “RecNo” is the position of the record in the table 1000. Thus, in the value order view, the item “RecNo” is added to each record. As a result, the position of the record in the table 1000 on which the value order view is based is also managed in the value order view. Therefore, for example, when a desired record is selected from the value order view by the user of the terminal device 10, the display processing unit 100 is based on the value order view based on the value of the item “RecNo” of the selected record. The corresponding record can be acquired from the table 1000.

For example, “1” is added as the value of the item “RecNo” to the record at position “0” in the value order view VW2-1 shown in FIG. This indicates that the record at position “0” in the value order view VW2-1 shown in FIG. 5A is the record at position “1” in the table 1000-1 shown in FIG. Yes. The same applies to the value of the item “RecNo” of other records.

In the value order view VW2-1 and the value order view VW2-2, the value of the item “Name” is displayed in alphabetical order (ascending alphabetical order).

In this way, the value order view is a view that displays a specified item of each record constituting the table 1000 in the order of the value of the item (for example, ascending order (or descending order) or alphabetical order). 5A and 5B show the value order view regarding the item “Name”, the value order view regarding the item “Age” can be considered in the same manner.

≪Record order UNION view≫
FIG. 6 shows a record order UNION view VW3 of each record constituting the table 1000-1 shown in FIG. 3A and each record constituting the table 1000-2 shown in FIG.

As shown in FIG. 6, in the record order UNION view VW3, each record configuring the table 1000-2 is displayed after each record configuring the table 1000-1. At this time, in the record order UNION view VW3, the position of each record constituting the table 1000-2 is reread. Specifically, for a record constituting the table 1000-2, when the position of the record is m and the number of records constituting the table 1000-1 is N, in the record order UNI view VW3, the position of the record is “ m + N ”. As a result, the records of positions “0” to “3” in the table 1000-2 shown in FIG. 3B are read as positions “4” to “7” in the record order UNI view VW3 shown in FIG. It has been.

In the record order UNI view VW3 shown in FIG. 6, each record constituting the table 1000-2 having the table number “1” is displayed after each record constituting the table 1000-1 having the table number “0”. However, this order may be reversed.

As described above, the record order UNION view is a view that displays the records constituting each table 1000 in the order of the table number and in the order of the records. In FIG. 6, the record order UNION view of two tables 1000 is shown, but the record order UNION view of three or more tables 1000 can be considered in the same manner.

≪UNION view in order of value≫
FIG. 7 shows a value order UNION view VW4 regarding the item “Name” of each record constituting the table 1000-1 shown in FIG. 3A and each record constituting the table 1000-2 shown in FIG. 3B. It is.

As shown in FIG. 7, an item “RecNo” is added to each record of the value order UNION view VW4. The item “RecNo” is as described above, but the position of each record constituting the table 1000-2 can be read. Specifically, in the same manner as described above, regarding the record constituting the table 1000-2, when the position of the record is m and the number of records constituting the table 1000-1 is N, the record order UNION view VW3 The position of the record is read as “m + N”.

Therefore, for example, when the user of the terminal device 10 selects a desired record from the value order UNION view, the display processing unit 100 displays the value order UNION view based on the value of the item “RecNo” of the selected record. The corresponding record can be acquired from the base table 1000. Details of this method will be described later.

Also, in the value order UNION view VW4, the value of the item “Name” is displayed in alphabetical order (ascending alphabetical order). More specifically, in the value order UNION view VW4, items “Name” values, table numbers, and positions in the table 1000 are displayed in order from the highest priority.

In this way, the value order UNION view displays a specified item of each record constituting each of the plurality of tables 1000 in the order of the value of the item (for example, ascending order (or descending order) or alphabetical order). It is a view. In FIG. 7, the value order UNION view regarding the item “Name” is shown, but the value order UNION view regarding the item “Age” can be considered similarly.

<Internal data structure>
Next, an internal data structure of the database 300 of the data management apparatus 20 will be described. As described above, the database 300 stores SVL 1100, IND 1200, and INV 1300 as internal data created from the table 1000. Therefore, hereinafter, regarding the table 1000-1 and the table 1000-2 shown in FIGS. 3A and 3B, the SVL 1100, the IND 1200, and the INV 1300 created from these tables 1000 will be described.

FIG. 8A shows a table 1000-1 having a table number “0” and SVL1100-1, IND1200-1 and INV1300-1 created for the item “Name” of each record constituting the table 1000-1. Is shown.

FIG. 8B shows a table 1000-2 having a table number “1” and SVL1100-2, IND1200-2, and INV1300- created for the item “Name” of each record constituting the table 1000-2. 2. Here, when rereading of the position of the record occurs (that is, when unioning with another table 1000), the position of INV1300-2 is also reread. That is, for example, when the positions “0” to “3” of the table 1000-2 are replaced with the positions “4” to “7”, the positions “0” to “3” of the INV 1300-2 are also respectively illustrated as shown in the figure. It is read as “7”, “5”, “4”, “6”.

The SVL 1100 shown in FIG. 8A and FIG. 8B shows the maximum value at the end after setting the value of the item “Name” of each record constituting the table 1000 to unique and ascending order (ascending order of the alphabet). The value “∞” is added.

IND 1200 shown in FIGS. 8A and 8B stores the start position of INV 1300 corresponding to each element of SVL 1100. For a certain element of the SVL 1100, the position from the start position corresponding to the element to the one before the start position corresponding to the next element of the SVL 1100 indicates the corresponding position of the INV 1300 for the certain element. For example, for “Cathy” of SVL1100-1, the corresponding element of IND1200-1 is “1”. On the other hand, for the element “Emily” next to “Cathy” of the SVL 1100-1, the corresponding element of the IND 1200 is “3”. Therefore, for “Cathy” of the SVL 1100, the position of the corresponding element of the INV 1300-1 is “1” to “2”.

Further, each element of IND1200 gives the number of appearances of a value smaller than SVL1100 corresponding to the element. For example, the element of IND1200-1 corresponding to “Emily” of IND1200-1 shown in FIG. 8A is “3”. This indicates that the number of records in which the value of the item “Name” is smaller than “Emily” is “3” in the table 1000-1.

INV1300 shown in FIGS. 8A and 8B is a transposed record number. For example, in the INV 1300-1 shown in FIG. 8A, the element corresponding to “Emily” in the SVL 1100-1 is “2”. This indicates that in the table 1000-1, the record whose item “Name” is “Emily” is stored at the position “2”. Similarly, for example, in the INV 1300-1 shown in FIG. 8A, the elements corresponding to “Cathy” of the SVL 1100-1 are “0” and “3”. This indicates that in the table 1000-1, a record whose item “Name” is “Cathy” is stored at positions “0” and “3”.

As shown in FIG. 8, SVL 1100, IND 1200, and INV 1300 are realized by array data, lists, vector data, and the like.

8 shows SVL1100, IND1200, and INV1300 created for the item “Name”, but FIG. 9 shows SVL1100, IND1200, and INV1300 created for the item “Age” as another example.

FIG. 9A shows a table 1000-1 having a table number “0” and SVL1100-1, IND1200-1 and INV1300-1 created for the item “Age” of each record constituting the table 1000-1. Is shown. FIG. 9B shows the table 1000-2 with the table number “1” and the SVL1100-2, IND1200-2 and INV1300- created for the item “Age” of each record constituting the table 1000-2. 2. Note that, when the rereading of the position of the record occurs in FIG. 9B, the position of the INV 1300-2 is also reread as described above. That is, for example, when the positions “0” to “3” of the table 1000-2 are replaced with the positions “4” to “7”, the positions “0” to “3” of the INV 1300-2 are also respectively illustrated as shown in the figure. It is read as "5", "7", "4", "6".

In FIG. 9, the values of the elements of SVL 1100, IND 1200, and INV 1300 are different from those in FIG. 8, but the meanings represented by these data are the same.

<Method for creating SVL1100, IND1200, and INV1300>
Next, as an example, the case where the SVL 1100-1, IND 1200-1, and INV 1300-1 shown in FIG. 8A are created from the table 1000-1 shown in FIG. 3A will be described with reference to FIG. To do. Each process of S1 to S5 in FIG. 10 is performed by the data processing unit 200.

S1. Add RecNo: The data processing unit 200 extracts each element of the item “Name” from the table 1000-1 and adds an item “RecNo” having a record number as a value to each element. At this time, the record number may be given the same number as the position of each element.

S2. Sort By Value: The data processing unit 200 sorts by the value of the item “Name” (for example, sorts in ascending order).

S3. Separate: The data processing unit 200 separates the item “Name” and the item “RecNo” for each element. Each element of the item “RecNo” after the separation becomes INV1300.

S4. Unique & Count: The data processing unit 200 creates an item “Count” whose value is the number of occurrences of each value of the item “Name” and eliminates the same value from the item “Name”.

S5. Aggregate: The data processing unit 200 adds an element whose value is “0” to the head of the item “Count”, and then converts the number of appearances into a cumulative number. Further, the data processing unit 200 adds an element whose value is “∞” to the end of the item “Name”. Each element of the item “Name” after performing these processes is SVL1100, and each element of the item “Count” is IND1200.

Note that the number of occurrences is converted into a class count when each element of the item “Count” is c [i] (i = 1 = 1,..., I), in order from j = 2 to j = I. c [1] +... + c [j] is changed to c [j].

In the example shown in FIG. 10, as an example, the SVL 1100, IND 1200, and INV 1300 related to the item “Name” are created.

Note that, in the processing of S1 to S5, for example, when a new table 1000 is added to the database 300, or a new record is added to the table 1000 in which SVL1100, IND1200, and INV1300 have already been created. Done in some cases.

Further, in the present embodiment, it has been described that the above-described processes of S1 to S5 are all performed by the data processing unit 200. However, the present invention is not limited to this. It may be executed by a device different from the device 20. In particular, after all the processing of S1 to S5 is executed by a device different from the data management device 20, the processing result of this device may be stored in the data management device 20.

<Search method>
Here, when the value order view or the value order UNION view is obtained, when searching for desired data from the value order view or the value order UNION view, the search is performed using the known bisection method (binary search method). can do. The data search can be performed, for example, by the user of the terminal device 10 specifying a search condition (for example, =, <,>, between, negation of these, etc.). Note that the data search may be performed by the display processing unit 100 or the data processing unit 200.

At this time, in the present embodiment, for the following two reasons, desired data can be searched at a high speed from the value order view or the value order UNION view, and the search result can be held with a small data capacity.

Reason 1: It is possible to specify the start position and end position of a hit area (hit area) by a known bisection method or the like. For this reason, it is not necessary to evaluate the search condition for records other than the record at the head position and the record at the tail position.

Reason 2: When the number of hits is enormous, in the conventional technique, it is necessary to hold all records in the hit area as a search result. In this embodiment, the start position and the end of the hit area are included as a search result. It is sufficient to hold only the position.

The reason 2 described above will be described in more detail. For example, in the value-ordered UNION view VW4 shown in FIG. 7, when searching for records whose value of the item “Name” is “Cathy” to “Dorothy”, the corresponding record Are records at positions “2” to “6”. For this reason, in this embodiment, it is sufficient to hold only the head position “2” and the tail position “6” of the hit area. Here, since the length of the hit area is 6-2 + 1 = 5 in this case, the number of hits is 5. Therefore, hereinafter, a case where all five records are extracted will be described.

Note that before this explanation, the value order UNION view VW4 shown in FIG. 7 is virtual. Therefore, in order to retrieve each record from the value order UNI view VW4 shown in FIG. 7, it is necessary to call a routine for retrieving a specified row of the value order UNI view VW4. Since “RecNo” returned from this routine has been read, the table 1000 is specified by determining whether it is the table 1000-1 or the table 1000-2 from the value of RecNo, and the read is changed. Return RecNo as a result of It should be noted that the RecNo after the replacement is returned, the table 1000 is used to identify it, and the routine to return the RecNo before the replacement is performed, for example, without following the procedure of returning to the RecNo before the replacement. It may be used.

(1) The RecNo of the first record hit is 0. RecNo = 0 belongs to the table 1000-1 (because it is less than 4). The offset of table 1000-1 = 0 is subtracted from RecNo = 0, and RecNo = 0-0-0 = 0 in the table. Therefore, a record at position “0” in the table 1000-1 is obtained.

(2) The RecNo of the second record hit is 3. RecNo = 3 belongs to the table 1000-1 (because it is less than 4). The offset of table 1000-1 = 0 is subtracted from RecNo = 3, and RecNo = 3-0 = 3 in the table. Therefore, a record at position “3” in the table 1000-1 is obtained.

(3) The RecNo of the third record hit is 5. RecNo = 5 belongs to the table 1000-2 (because it is 4 or more). The offset of the table 1000-2 = 4 is subtracted from RecNo = 5, and RecNo = 5-4 = 1 in the table. Therefore, a record at position “1” in the table 1000-2 is obtained.

(4) The RecNo of the fourth record hit is 4. RecNo = 4 belongs to the table 1000-2 (because it is 4 or more). The offset of table 1000-2 = 4 is subtracted from RecNo = 4, and RecNo = 4-4 = 0 in the table. Therefore, a record at position “0” in the table 1000-2 is obtained.

(5) The RecNo of the fifth record hit is 6. RecNo = 6 belongs to the table 1000-2 (because it is 4 or more). The offset of table 1000-2 = 4 is subtracted from RecNo = 6, and RecNo = 6-4 = 2 in the table. Therefore, a record at position “2” in the table 1000-2 is obtained.

Note that the view representing the search result may be a view in which the records in the hit area are displayed among the records displayed in the value order view or the value order UNION view. A view representing the search result is displayed by the display processing unit 100.

<How to create each view>
Next, a method for creating each view will be described. By creating each view, the display processing unit 100 can display these views. The display processing unit 100 may create and display a view by acquiring data from the data management device 20, or the data processing unit 200 may generate and create a view by acquiring data from the database 300. The display processing unit 100 may display the view by transmitting the processed view to the display processing unit 100. However, for example, when creating a view in which the tables 1000 managed by different data management apparatuses 20 are unioned, the display processing unit 100 preferably creates the view.

≪How to create record order view≫
For example, when creating the record order view VW1-1 of the table number “0”, the records from the positions “0” to “3” of the table 1000-1 shown in FIG. Similarly, for example, when creating the record order view VW1-2 of the table 1000-2 with the table number “1”, the records from the positions “0” to “3” of the table 1000-2 shown in FIG. Should be acquired in order.

≪How to create a value order view≫
For example, when creating the value order view VW2-1 for the item “Name” of the table 1000-1 with the table number “0”, each of the positions “0” to “3” of the INV 1300-1 shown in FIG. A record having the element value as its position may be acquired from the record order view VW1-1. More specifically, the value of each element at positions “0” to “3” of INV 1300-1 shown in FIG. 8A is INV [0] = 1, INV [1] = 0, INV [2]. = 3, INV [3] = 2. Therefore, the records at the positions “1”, “0”, “3”, “2” are obtained in order from the record order view VW1-1 shown in FIG. Each is added as a value of “RecNo”. As a result, the value order view VW2-1 shown in FIG. 5A is obtained.

Similarly, for example, when creating the value order view VW2-2 regarding the item “Name” of the table 1000-2 with the table number “1”, the positions “0” to “3” of the INV 1300-2 shown in FIG. It is only necessary to acquire a record having the position of each element value of “” from the record order view VW1-2. More specifically, the values of the elements at positions “0” to “3” of INV1300-2 shown in FIG. 8B are INV [0] = 3, INV [1] = 1, INV [2]. = 0 and INV [3] = 2. Therefore, the records at the positions “3”, “1”, “0”, “2” are sequentially acquired from the record order view VW1-2 shown in FIG. 4B, and these values are set as items for the acquired records. Each is added as a value of “RecNo”. As a result, the value order view VW2-2 shown in FIG. 5B is obtained.

≪How to create a record order UNION view≫
For example, when creating the record order UNION view VW3 shown in FIG. 6, the records from the positions “0” to “3” in the table 1000-1 shown in FIG. The records from position “0” to “3” in the table 1000-2 shown in FIG.

However, when the positions “4” to “7” after the replacement are designated when the records of the positions “0” to “3” in the table 1000-2 shown in FIG. The positions “4” to “7” are converted to the positions “0” to “3” before the replacement.

≪How to create a UNION view by value order≫
For example, when the value order UNION view VW4 shown in FIG. 7 is created, the items “" from the value order views VW2-1 and VW2-2 shown in FIG. 5A and FIG. The “Name” value, the table number, and the record number in the table may be acquired in this order.

Therefore, in this case, the records are acquired from the value order view VW2-1 shown in FIG. 5A and the value order view VW2-2 shown in FIG. 5B in the following order.

(1) Record at position “0” in the value order view VW2-1 shown in FIG. 5A (2) Record at position “0” in the value order view VW2-2 shown in FIG. 5B (3) FIG. Record of position “1” in the value order view VW2-1 shown in FIG. 5 (a) (4) Record of position “2” in the value order view VW2-1 shown in FIG. 5A (5) FIG. Record of position “1” in value order view VW2-2 shown in (6) Record of position “2” in value order view VW2-2 shown in FIG. 5B (7) Value order shown in FIG. 5B Record of position “3” in view VW2-2 (8) Record of position “3” in value order view VW2-1 shown in FIG. 5A At this time, (2), (5) to (7) above Then, the item “RecNo” is replaced.

As described above, the data management system 1 according to the present embodiment uses the SVL 1100, the IND 1200, and the INV 1300 when creating and displaying various views, so that even if the number of records constituting the table 1000 is enormous. Various views can be created and displayed easily and quickly. In addition, the data management system 1 according to the present embodiment uses the SVL 1100, the IND 1200, and the INV 1300 when performing a data search, so that the search can be performed at a high speed and the search result can be obtained with a small data capacity. Will be able to hold.

In the present embodiment, a view in which a plurality of tables 1000 are UNION has been described. However, JOIN can be used instead of UNION. In this case, for example, S4. When calculating the number of occurrences of the value of the corresponding item in Unique & Count, the number of occurrences may be multiplied by the number of occurrences of the matching value in the table 1000 to be joined, to create IND1200 and INV1300.

<Various functions>
Here, various functions defined for the table 1000 having the above data structure will be described.

-Count_LT (table number, item name, value)
This function obtains the number of records having a value smaller than the specified value for the item with the specified item name in the table with the specified table number.

Count_LT, for example, searches the SVL 1100 related to the item of the specified item name, and if there is an element that matches the specified value, the position of that element may be i and IND [i] may be acquired. On the other hand, Count_LT, for example, searches SVL 1100 related to the item of the specified item name, and if there is no element that matches the specified value, the value of each element of SVL 1100 is larger than the specified value. In addition, IND [i] may be obtained by setting i as the position of the element having the minimum value.

-Count_LE (table number, item name, value)
This function obtains the number of records with a value less than or equal to the specified value for the item with the specified item name in the table with the specified table number.

Count_LE, for example, searches the SVL 1100 related to the item of the specified item name, and if there is an element that matches the specified value, the position of the element may be i and IND [i + 1] may be acquired. On the other hand, Count_LE, for example, searches SVL 1100 for the item of the specified item name, and if there is no element that matches the specified value, the value of each element of SVL 1100 is greater than the specified value. In addition, IND [i] may be obtained by setting i as the position of the element having the minimum value.

-Count_EQ (table number, item name, value)
This function obtains the number of records having a value equal to a specified value for an item having a specified item name in a table having a specified table number.

Count_EQ may be, for example, Count_EQ () = Count_LE () −Count_LT ().

・ Pos (table number, item name, value)
This is a function for obtaining the position where the specified value appears in the SVL 1100 for the item of the specified item name in the table of the specified table number.

For example, Pos searches the SVL 1100 related to the item of the specified item name, and if there is an element that matches the specified value, the position of this element may be acquired.

・ Total_Count_LT (item name, value)
This function obtains the sum of all the tables having a record number smaller than the specified value for the item with the specified item name.

Total_Count_LT may be ΣCount_LT (Table [i], item name, value), with the table number as Table [i].

・ Total_Count_LE (item name, value)
This function obtains the sum of all the tables with the number of records below the specified value for the item with the specified item name.

Total_Count_LE may be ΣCount t_LE (Table [i], item name, value) with the table number as Table [i].

<Method of obtaining record number of record order view from value order UNION view>
As described above, the item “RecNo” is added to each record of the value order UNION view. Therefore, when the user of the terminal apparatus 10 selects a desired record from the value order UNION view, the position (record number) of the record order view of the corresponding table 1000 is determined from the value of the item “RecNo” of the selected record. If it can be obtained, a corresponding record can be acquired from the table 1000 by this position (record number).

Henceforth, using the function defined above, from the record number of the value order UNION view (value of the item “RecNo”), the record number of the record order view regarding the table 1000 and the table 1000 of this table number (the head) A method for obtaining the relative position from the record will be described. Note that the method described below may be executed by either the display processing unit 100 or the data processing unit 200 to which the record number r in the value order UNION view is given. Hereinafter, it is assumed that the SVL 1100 relates to the item “Name”. Therefore, hereinafter, the argument “item name” of the above function is assumed to be “Name”, and the description thereof is omitted. Specifically, for example, when the value of the item “Name” is v, Total_Count_LT (Name, v) is simply expressed as Total_Count_LT (v).

S11) Find v to which r belongs.

This can be done by searching for v that satisfies the following.

Total_Count_LT (v) ≦ r <Total_Count_LE (v)
Since each SVL 1100 is in ascending order and its value is unique, the above v can be efficiently searched using, for example, a known bisection method (binary search method).

As an example, for the item “Name” in the tables 1000-1 and 1000-2 shown in FIGS. 3A and 3B, the values of v for each r are shown in Table 1 below.

S12) A table number j is obtained from r and v.

As can be seen from Table 1 above, Total_Count_LT (v) gives the position where the value v starts in the value order UNION view. Therefore, when offset = r−Total_Count_LT (v), offset indicates how many records of the value r are lower than the start position of the value v in the value order UNION view.

Therefore, the table number j can be obtained by obtaining the minimum j satisfying the following expression 1.

As an example, Table 2 in which the table number j obtained by the above Equation 1 is added to the above Table 1 is shown below.

S13) The record number r0 of the record order view is obtained from r, v and j.

First, assuming that the number of all records in the table 1000 of each table number i satisfying i <j is “number of records 1”, the number of records 1 can be obtained by the following Expression 2.

Further, assuming that the number of all records in the table 1000 of each table number i satisfying i> j is “number of records 2”, the number of records 2 can be obtained by the following Expression 3.

Therefore, if rj = (r−number of records 1−number of records 2) is obtained, the position of the table number j in the value order view regarding the table 1000 can be obtained. Therefore, the record number of this table 1000 can be acquired by INV [rj] of the table 1000 of the table number j. The record number is the same as the relative position from the first record in the record order with respect to the table 1000.

If necessary, the record number INV [rj] can be converted into the record number r0 of the value order UNIO view by the following method. In other words, the record number INV [rj] of the j-th table 1000 (table 1000 with the table number j) that is the basis of the value order UNION view is the record number constituting the 0th to (j−1) -th table 1000. By adding the sum of the numbers, it can be converted into a record number (value of the item “RecNo”) in the value order UNION view.

As described above, the position (record number) of the record order view regarding the corresponding table 1000 can be obtained from the value (record number) of the item “RecNo” of the value order UNION view. Thereby, when a desired record is selected from the value order UNION view by the user of the terminal device 10, the selected record is acquired from the corresponding table 1000 (the table 1000 in which the selected record is stored). Will be able to.

<Application example 1>
Next, application example 1 of the present embodiment will be described. In application example 1, when displaying a value-ordered UNION view for a certain item, if there is a record having the same value of the item in the same table 1000, only one record is displayed for each table 1000 Will be explained.

For example, when displaying a value-ordered UNION view related to the item “Name”, as shown in FIG. 11, in the value-ordered view of each table 1000, records other than one of the records having the same value of the item “Name” After deletion, a value-ordered UNION view may be created.

In the example shown in FIG. 11, there are two records whose item “Name” is “Cathy” in the value order view of the table 1000-1 with the table number “0”. For this reason, of these two records, other than one record is deleted.

Similarly, in the value order view of the table 1000-2 with the table number “1”, there are two records whose item “Name” is “Dorothy”. For this reason, of these two records, other than one record is deleted.

Here, it is possible to arbitrarily decide which records of the same value are left without being deleted. For example, among records having the same value, the record with the smallest position in the value order view may be left, or the record with the smallest value (record number) of the item “RecNo” may be left.

Note that, as shown in Application Example 1, when only one of the records having the same value for each table 1000 is displayed in the value order UNION view, the Count_LT and the Count_LE defined above are corrected as follows.

-Count_LT (table number, item name, value)
For example, Count_LT searches the SVL 1100 related to the item of the specified item name, and if there is an element that matches the specified value, the position of the element is i and i can be obtained. On the other hand, Count_LT, for example, searches SVL 1100 related to the item of the specified item name, and if there is no element that matches the specified value, the value of each element of SVL 1100 is larger than the specified value. In addition, the position of the element having the minimum value may be i and i may be acquired.

-Count_LE (table number, item name, value)
For example, Count_LE searches the SVL 1100 related to the item of the specified item name, and if there is an element that matches the specified value, the position of the element is i and i + 1 may be acquired. On the other hand, Count_LE, for example, searches the SVL 1100 related to the item of the specified item name, and if there is no element that matches the specified value, the value of each element of the SVL 1100 is larger than the specified value. In addition, the position of the element having the minimum value may be i and i may be acquired.

<Application example 2>
Next, application example 2 of the present embodiment will be described. Application example 2 is an example in which the present embodiment is applied to keyword search.

For example, as shown in FIGS. 12A and 12B, each record constituting the table 1000-1 and the table 1000-2 includes an item “Text”, an item “Keyword”, and an item “RecNo”. It shall have. Here, the item “Text” is a column in which text including a keyword to be searched is stored. The item “Keyword” is a keyword extracted in advance from the text stored in the item “Text”.

At this time, as described in the present embodiment, by creating SVL 1100, IND 1200, and INV 1300 for the item “Keyword”, the text including the specified keyword among the texts stored in the item “Text”. You will be able to search. In addition, for example, since the value order UNION view of the table 1000-1 and the table 1000-2 can be displayed, the keyword search using the plurality of tables 1000 can be performed easily and at high speed.

For example, when a text including the keyword “car” is to be searched from the value order view of the table 1000-1, INV [3] = 1, INV [1] is used using SVL1100-1, IND1200-1, and INV1300-1. 4] = 2 and INV [5] = 3 are obtained. As a result, the record with the record number “1”, the record with the record number “2”, and the record with the record number “3” can be acquired from the table 1000-1 as text including the designated keyword. . The same applies to the keyword search in the value order UNION view of the table 1000-1 and the table 1000-2.

The present invention is not limited to the specifically disclosed embodiments, and various modifications and changes can be made without departing from the scope of the claims.

This application claims the priority of the basic application 2018-028665 filed on February 21, 2018 with the Japan Patent Office, the entire contents of which are incorporated herein by reference.

DESCRIPTION OF SYMBOLS 1 Data management system 10 Terminal device 20 Data management apparatus 100 Display processing part 200 Data processing part 1000 Table 1100 SVL
1200 IND
1300 INV

Claims (9)

1. A table storing a plurality of records; a first array data storing predetermined item values included in each of the plurality of records; a second array data storing a record number for identifying the records; Storage means for storing third array data for associating the first array data with the second array data;
   Of the plurality of records stored in the table, using the first array data, the second array data, and the third array data in accordance with the designation of the item value or the record number, Obtaining means for obtaining one or more records corresponding to the specified item value or the record number;
   An information processing system comprising:
2. 2. The information processing according to claim 1, wherein the first array data is data in which predetermined item values included in each of the plurality of records are stored in an array without duplication in the order of the item values. system.
3. The information processing system according to claim 1 or 2, wherein the second array data is data in which the record numbers are stored in the array in the order of the item values.
4. The third array data is data in which an accumulated number obtained by accumulating the number of appearances of the item value in the order of records stored in the table is stored in the array. Information processing system.
5. The information processing system includes a plurality of data management devices each having the storage means,
   The acquisition means includes
   The first array data, the second array data, and the third array data stored in the storage unit respectively included in the plurality of data management devices according to the designation of the item value or the record number , Among the plurality of records stored in the table stored in the storage unit respectively included in the plurality of data management devices, one or more corresponding to the specified item value or the record number The information processing system according to claim 1, wherein a record is acquired.
6. The acquisition means includes
   The first array data and the second array data corresponding to the second table according to the designation of the record number in the view in which the first table and the second table are combined in the order of the item values. 6. One or more records corresponding to the designated record number are obtained from the second table by rereading the record number using the third array data and the third array data. Information processing system described in 1.
7. The acquisition means includes
   The information processing system according to claim 6, wherein the record number is reread by subtracting the number of records in the first table from the record number.
8. The acquisition means includes
   Of the one or more records corresponding to the specified item value or the record number, the first information specifying the first record and the second information specifying the last record are acquired and acquired. The information processing system according to claim 1, wherein the first information and the second information are stored in a predetermined storage area as a search result.
9. A table storing a plurality of records; a first array data storing predetermined item values included in each of the plurality of records; a second array data storing a record number for identifying the records; An information processing apparatus having storage means for storing third array data that associates the first array data with the second array data,
   Of the plurality of records stored in the table, using the first array data, the second array data, and the third array data in accordance with the designation of the item value or the record number, An acquisition procedure for acquiring one or more records corresponding to the specified item value or the record number;
   The information processing method characterized by performing.
   

Images