WO2018080187A1

WO2018080187A1 - Apparatus for operating multiple databases of embedded database system, and method therefor

Info

Publication number: WO2018080187A1
Application number: PCT/KR2017/011895
Authority: WO
Inventors: 김재광; 한혁; 김동일; 진성일
Original assignee: 주식회사 리얼타임테크; 현대엠엔소프트 주식회사
Priority date: 2016-10-26
Filing date: 2017-10-26
Publication date: 2018-05-03
Also published as: KR101828328B1

Abstract

The present invention relates to an apparatus for operating multiple databases of an embedded database system, and a method therefor, the apparatus comprising: a plurality of databases into which a data page for storing data is loaded, and which grants a page ID to the data page; a connection manager for performing database access through a dynamic management including generation or removal of a database ID for each database when the at least one database is being operated; a plurality of query processors for generating a page request instruction including the database ID and the page ID so as to perform, on the data page, query processing including insertion, deletion, change, and retrieval of data; and a buffer manager, which selects a corresponding database from among the plurality of databases by using the database ID according to the page request instruction, and reads the data page from the selected database, by using the page ID, so as to cache and manage the data page.

Description

Multi-database operating device and method for embedded database system

The present invention relates to an apparatus and method for operating multiple databases of an embedded database system, and more particularly, to an apparatus and method for operating multiple databases of an embedded database system capable of operating multiple databases simultaneously using a single buffer manager. It is about.

A database system is a database overall management system, and a database management system is a major component of a database system that performs a user's data processing needs while managing a database. Such a database management system is located between a user and a database, and can be classified into a query processor and a storage data manager (or buffer manager) according to functions.

1 is a view for explaining an embedded database management system according to the prior art.

Referring to FIG. 1, the embedded database management system includes a database 1, a connection manager 10, a query processor 20, a buffer manager 30, and a lower module 40.

The database 1 is a file in which data pages are stored, and the connection manager 10 is in charge of database operation.

The query processor 20 is responsible for interpreting and processing a user's data processing request, and the buffer manager 30 manages a user database and a data dictionary (information about data stored in the database) stored on disk. In this case, it plays the role of actually accessing.

That is, the buffer manager 30 reads the data page loaded in the database 1 or writes to the database 1 when the data page already read is modified, and influences the performance of the database system. It performs data page cache and management.

The embedded database management system is composed of an upper module and a lower module based on the buffer manager 30. The upper module is a module using the buffer manager 30, and the lower module is a module used by the buffer manager 30. Therefore, the query processor 20 corresponds to a higher module.

Since such an embedded database management system requires a buffer manager 30 for each database, three buffer managers 300 are required when operating three databases 1 at the same time. In the case where the buffer pool memory of 30) is 32MB, it can be seen that the buffer pool memory of the buffer manager required to operate three databases 1 simultaneously consumes 96MB in total.

On the other hand, the data programming into the multi-plane flash memory of Korea Patent Publication No. 2011-0069049 is a data transfer scheme in which data pages are programmed to allow multiple data pages to be programmed without increasing the size of the page buffers. Allow it to be controlled.

Conventional data programming into a multi-plane flash memory includes a number of flash memory devices and is connected to a page buffer through a memory interface, using as many page buffers as the number of flash memory devices.

As such, the embedded database management system of the related art generally cannot operate several databases at the same time. However, when operating several databases at the same time, a buffer manager is required as many as the number of databases, which consumes a lot of memory resources. There is a problem.

The present invention can operate multiple databases at the same time using only one buffer manager, which is an essential configuration for the performance of the embedded database system with limited memory resources, and thus the multiple database operating apparatus of the embedded database system, which can greatly reduce memory usage. It provides a way.

Among the embodiments, the multiple database operating apparatus of the embedded database system includes a plurality of databases loaded with data pages for storing data and assigning page IDs to the data pages; A connection manager that processes database access through dynamic management, including creation or destruction of a database ID for each database when at least one database is running; A plurality of query processors for generating a page request command including the database ID and the page ID to perform a query processing operation including inserting, deleting, changing, and retrieving data in the data page; And a buffer manager that selects a corresponding database among a plurality of databases using the database ID according to the page request command, reads the data page using the page ID from the selected database, and caches and manages the data page. It is characterized by including.

The connection manager further includes a database ID manager that uses an array data structure, sets the array size to the maximum number of databases that can be simultaneously connected, and manages database IDs for distinguishing databases using array cursors. It is characterized by.

The database ID manager uses the array cursor as an unsigned short type, sets the range of the array cursor value to 0 to 65535, and includes a statement statement including close and open to control the array cursor. It characterized in that to use.

The database ID manager, when the array is configured with as many fields as the preset array size, inputs the array cursor by '1' until the database open command is input and an empty field is output from the array. If an empty field is found, the array cursor located in the corresponding field is assigned to a database ID and a database connection is performed. If the database close command is input, the array cursor is located in a field corresponding to the database ID. The database connection is terminated after initializing the field in which the array cursor is located by using a remaining operation using a database ID and an array size.

The database ID manager outputs a message indicating that the maximum number of database connections has been exceeded when the database Open command is input but there are no empty fields in the array, and returns a database connection failure.

A plurality of submodules are located between the buffer manager and the database to use a lower database using database instructions that perform one or more logical functions, and the buffer manager may be configured in an application programming interface (API). And a submodule manager configured to add a parameter, select one submodule among a plurality of submodules using the database ID, and read a data page through the selected submodule.

The connection manager uses an array data structure and sets the array size to the maximum number of databases that can be connected at the same time. The submodule manager uses a remaining operation using a database ID and an array size to select one of a plurality of submodules. Selecting a submodule to use.

The buffer manager caches a data page by displaying a database ID and a page ID in an internal buffer pool memory, and reads the data page from the buffer pool memory using the page ID and the database ID when re-requesting the data page. It is characterized by coming.

Among the embodiments, the multiple database operating method of the embedded database system, connection manager for processing database access to a plurality of databases loaded with data pages, query processing including the insertion, deletion, change, search of data in the data page In a multi-database operating method of an embedded database system including a query processor that performs a task and a buffer manager that reads a data page from the database and caches and manages the data page, the connection manager is running in a database. Processing a database access through dynamic management of a database ID for creating or destroying a database ID for distinguishing a plurality of databases to use one buffer manager). The query processor requesting a page request command to a buffer manager using a database ID and a page ID of a data page, and performing a query processing operation including inserting / deleting / modifying / searching data into the requested data page; And the buffer manager adds a parameter of a database ID, selects a corresponding database among a plurality of databases using the database ID according to the page request command, and selects a data page corresponding to a page ID from the selected database. And reading and performing the role of caching and managing the data page in the buffer pool memory.

The connection manager may process a database access through dynamic management of a database ID for generating or destroying a database ID for distinguishing a plurality of databases in order to use one buffer manager, wherein the array is a field corresponding to a preset array size. When the database Open command is input, the array cursor is set to '1' to check whether the database is connected to the field where the current array cursor is located and to search for an empty field when the database is connected to the field. Increasing; And if there is an empty field in the array, the database cursor is assigned after assigning the value of the array cursor located in the field as the database ID, and if the empty field is not present in the array, the maximum number of database connections is exceeded. Outputting a message informing of the failure, and returning a database connection failure.

The connection manager may process a database access through dynamic management of a database ID for generating or destroying a database ID for distinguishing a plurality of databases in order to use one buffer manager, wherein the array is a field corresponding to a preset array size. When the database close command is input, an array cursor is positioned in a field corresponding to a database ID, and a database connection is initialized after initializing a field in which the array cursor is located by using the remaining operations using the database ID and the array size. It is characterized by ending.

A plurality of submodules is located between the buffer manager and the plurality of databases to use a lower database using database instructions that perform one or more logical functions, and the buffer manager is located in an application programming interface (API). And a submodule manager configured to add an ID parameter, select one submodule among a plurality of submodules by using the database ID, and read a data page through the selected submodule. .

When the connection manager uses an array data structure and sets the array size to the maximum number of databases that can be simultaneously connected, the submodule manager uses a plurality of submodules by using the remaining operations using the database ID and the array size. It is characterized by selecting a sub module to be used.

The multi-database operating apparatus and method thereof of the embedded database system of the present invention can save multiple memory resources by simultaneously operating several databases using only one buffer manager in an embedded database system having limited memory resources. There is.

In addition, the present invention applies a database ID to operate multiple databases, manages the database ID using an array data structure and an array cursor, connect the database by designing the range of the array cursor value as wide as 0 ~ 65535 It is possible to prevent the buffer manager from malfunctioning without sacrificing performance on / shutdown.

In the present invention, when the database ID is operated, it is dynamically managed only in the program area and is not stored in the database. Therefore, there is no dependency between the plurality of databases. There is an effect that the operation of can be enabled.

2 is a diagram illustrating a configuration of an apparatus for operating multiple databases in an embedded database system according to an embodiment of the present invention.

3 is a diagram illustrating an operation state between a query processor and a buffer manager according to an exemplary embodiment of the present invention.

4 is a diagram illustrating a method for operating multiple databases of an embedded database system according to an embodiment of the present invention.

5 is a diagram illustrating a driving state of a database ID manager according to an embodiment of the present invention.

6 is a diagram illustrating a malfunction state of a buffer pool memory according to a range of array cursor values according to an embodiment of the present invention.

Description of the present invention is only an embodiment for structural or functional description, the scope of the present invention should not be construed as limited by the embodiments described in the text. That is, since the embodiments may be variously modified and may have various forms, the scope of the present invention should be understood to include equivalents capable of realizing the technical idea. In addition, the objects or effects presented in the present invention does not mean that a specific embodiment should include all or only such effects, the scope of the present invention should not be understood as being limited thereby.

On the other hand, the meaning of the terms described in the present invention will be understood as follows.

Terms such as "first" and "second" are intended to distinguish one component from another component, and the scope of rights should not be limited by these terms. For example, the first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being "connected" to another component, it should be understood that there may be other components in between, although it may be directly connected to the other component. On the other hand, when a component is referred to as being "directly connected" to another component, it should be understood that there is no other component in between. On the other hand, other expressions describing the relationship between the components, such as "between" and "immediately between" or "neighboring to" and "directly neighboring to", should be interpreted as well.

Singular expressions should be understood to include plural expressions unless the context clearly indicates otherwise, and terms such as "comprise" or "have" refer to features, numbers, steps, operations, components, parts, or parts thereof described. It is to be understood that the combination is intended to be present and does not exclude in advance the possibility of the presence or addition of one or more other features or numbers, steps, operations, components, parts or combinations thereof.

In each step, an identification code (e.g., a, b, c, etc.) is used for convenience of description, and the identification code does not describe the order of the steps, and each step clearly indicates a specific order in context. Unless stated otherwise, they may occur out of the order noted. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. Generally, the terms defined in the dictionary used are to be interpreted as being consistent with the meanings in the context of the related art, and should not be interpreted as having ideal or excessively formal meanings unless clearly defined in the present invention.

Referring to FIG. 2, a multi-database operating apparatus of an embedded database system may include a plurality of databases 100, a connection manager 200, a plurality of query processors 300, a buffer manager 400, and a plurality of submodules 500. It includes.

The plurality of databases 100 load data pages for storing data and assign page IDs to the data pages.

The connection manager 200 processes database access through dynamic management including generation and destruction of a database ID for each database 100 when the at least one database 100 is in runtime.

The connection manager 200 uses an array data structure, sets the array size to the maximum number of databases 100 that can be connected at the same time, and sets a database ID for identifying the database 100 using an array cursor. Database ID manager 210 that creates and manages.

Although the page ID is the same in the buffer pool memory of the buffer manager 400, data pages read from different databases 100 are mixed. Therefore, when using multiple databases 100 at the same time, the database ID manager 210 allocates and manages a database ID for each database 100 to distinguish which data page is read from which database 100.

The plurality of query processors 300 generate a page request command including a database ID and a page ID to perform a query processing operation including inserting, deleting, changing, and retrieving data using the page ID of the data page. In this case, the query processor 300, which is an upper module using the buffer manager 400, is provided as many as the number of lower modules 500 used by the buffer manager 400 to use one buffer manager 400.

The query processor 300 may manage the page ID and request a corresponding data page from the buffer manager 400 using the page ID when a data page associated with the page ID is needed. In addition, the query processor 300 notifies the buffer manager 400 that the data page has been modified after performing the modification such as inserting / deleting / changing data in the requested data page. Thus, the buffer manager 400 records the modified data page.

The buffer manager 400 implements by adding a parameter of a database ID to an application programming interface (API), selects a database 100 to be used using a database ID according to a page request command, and selects a database selected using a page ID. After reading the data page corresponding to the page ID at 100, the data page is cached and managed.

The data pages are cached in the buffer pool memory inside the buffer manager 400. The more the data page cache work, the better the database access performance, but the more buffer pool memory is consumed. Therefore, in order to operate three databases simultaneously, three 32MB buffer managers were required, so total 96MB of bufferpool memory was consumed. However, in the present invention, only one buffer manager 400 is required. Consumed.

In general, since most of the memory used by the embedded database system is occupied by the buffer pool memory, the buffer pool memory may be greatly reduced when one buffer manager 400 operates several databases at the same time.

The lower module 500 is installed between the buffer manager 400 and the database 100 and uses the lower database 100 by using database commands that perform one or more logical functions.

The buffer manager 400 selects one submodule 500 of the plurality of submodules 500 using a database ID, and reads the data page through the selected submodule 500. It includes.

Referring to FIG. 3, the buffer manager 400 implements by adding a parameter of a database ID to an API, and the query processor 300 adds a database ID together with a page ID to a page request command when using the buffer manager 400. do.

Accordingly, the buffer manager 400 performs remaining operations using the database ID and the array size to determine which submodule among the plurality of submodules 500 should be used using the database ID according to the page request command input from the query processor 300. The data page is read through the selected submodule 500.

At this time, the buffer manager 400 displays a database ID and a page ID to manage the read data page, and when a request for re-request of the data page is received, the buffer manager 400 uses the page ID and the database ID in the buffer pool memory. Find and return exactly.

4 is a diagram illustrating a method for operating multiple databases of an embedded database system according to an embodiment of the present invention, and FIG. 5 is a diagram illustrating a driving state of a database ID manager according to an embodiment of the present invention.

4 and 5, in a multi-database operating method of an embedded database system, a database ID for the connection manager 200 to separately operate a plurality of databases 100 in order to use one buffer manager 400. The manager 210 is included. (S1)

The database ID manager 210 uses an array cursor as an unsigned short type, sets the range of the array cursor to 0 to 65535, and commands such as Declare / Open / Fetch / Close to control the array cursor. Use sentences.

If the array size is 5, the array consists of 5 fields, and as shown in FIG. 5 (a), the connection manager 200 sets the array cursor to 0 to 4 when the database open command is input. Move the field in increments of one. If a database is not connected to the field where the array cursor is located and is empty, the value of the corresponding array cursor is assigned as the database ID. When the buffer manager 400 is used in the query processor 300, the parameter of the database ID is added and used.

If the database Open command is entered but all the fields in the array are filled, the database connection failure is returned with the message "The maximum number of database connections has been exceeded."

As shown in (b) of FIG. 5, the database ID manager 210 locates an array cursor in a field of array [1] when a Close command for a database having a database ID of '1' is inputted. Initialize the value of the field (array [1]) using the result of the rest of the operation using the array size) and terminate the database connection.

The database ID manager 210 dynamically generates / deletes the database ID only in the program area when the at least one database 100 is in operation. (S1) As described above, in the present invention, the database ID is a database ( 100, there is no dependency between a plurality of databases because they are not stored and managed, thereby enabling operations such as registering / unregistering / exchanging databases without interruption of the database system.

At this time, the array cursor is the minimum value (0) of the unsigned short and the maximum value of the unsigned short is 65535. Thus, the range of values for the array cursor is designed from 0 to 65535, and the range of values is repeated over and over until the database system shuts down.

The query processor 300 transmits a page request command to the buffer manager 400 using a database ID and a page ID, and performs a modification such as inserting / deleting / modifying data into a data page corresponding to the requested page ID. After that, the buffer manager 400 is notified that the data page has been modified.

The buffer manager 400 implements by adding a parameter of a database ID to the API, and selects a database 100 to be used among the plurality of databases 100 using the database ID according to the page request command (S3 and S4).

The buffer manager 400 reads the data page into the buffer pool memory 420 using the page ID in the selected database 100 to cache and manage the data page. (S5) At this time, the buffer manager 400 buffers the buffer pool. By displaying the page ID and database ID of the data page in memory, if the data page is requested again later, the page ID and database ID can be used to accurately find and return the data page.

As shown in FIG. 6, the database ID manager 210 generates a database ID using an array cursor value. When the array cursor value is narrow, the database ID manager 210 repeatedly executes an open / close command. In the bufferpool memory of), there is a possibility that there are data pages that have the same page ID but are read from different databases.

If the array size is 3 and the range of the array cursor value is 0 to 2, the range of the database ID is also 0 to 2.

A database ID [0] is generated according to the database open command, and the data page [0] corresponding to the page ID [0] is read from the database 100 into the buffer pool memory 420. Then, according to the database close command, disconnect the database whose database ID is '0' and initialize array [0]. At this time, the buffer manager 400 is too expensive to initialize the buffer pool memory of a relatively large storage capacity for each database Close command, and thus leaves the data page [0] as it is for performance.

In this manner, the buffer manager 400 reads data page [1] and data page [2] for the database with database IDs '1' and '2' according to the repetition of the database Open / Close command and then array [ 1] and array [2] are initialized, leaving data page [1] and data page [2] in the buffer pool memory.

After that, when the database open command is input, the array cursor is positioned at array [0], and the database ID is assigned to '0'. Therefore, instead of reading the data page [3] from the fourth database [3], the buffer manager 400 finds and returns the data page [0] remaining in the buffer pool memory, and returns the data page [0] from the buffer pool memory. The database [3] is not read because it is found.

As described above, in order to prevent a malfunction of the buffer pool memory generated when the range of the array cursor value is set narrow, the range of the value of the array cursor should be relatively large.

For example, in a database system, if the data page size is 16 KB and the buffer pool memory is 32 MB, up to 2048 data pages can exist in the buffer pool memory. Therefore, the database ID range is 0 to 65535. Malfunction of the buffer pool memory can be prevented.

In other words, since the database ID starts at 0 and passes through 65535 and becomes 0 again, the data page [0] for the database ID in the buffer pool memory disappears as a victim page, and thus finds and returns an invalid data page in the buffer pool memory. It can prevent.

Although described above with reference to a preferred embodiment of the present invention, those skilled in the art will be variously modified and changed within the scope of the invention without departing from the spirit and scope of the invention described in the claims below I can understand that you can.

Claims

A plurality of databases in which data pages for storing data are loaded, and a page ID is assigned to the data pages;

A connection manager that processes database access through dynamic management, including creation or destruction of a database ID for each database when at least one database is running;

A plurality of query processors for generating a page request command including the database ID and the page ID to perform a query processing operation including inserting, deleting, changing, and retrieving data in the data page; And

A buffer manager that selects a corresponding database from a plurality of databases using the database ID according to the page request command, and reads a data page using the page ID from the selected database to cache and manage the data page; Multiple database operating device of the embedded database system, characterized in that.
The method of claim 1,

The connection manager further includes a database ID manager that uses an array data structure, sets the array size to the maximum number of databases that can be simultaneously connected, and manages database IDs for distinguishing databases using array cursors. Multiple database operating device of the embedded database system, characterized in that.
The method of claim 2,

The database ID manager uses the array cursor as an unsigned short type, sets the range of the array cursor value to 0 to 65535, and includes a statement statement including close and open to control the array cursor. Multiple database operating device of the embedded database system, characterized in that using the.
The method of claim 3,

If the database ID manager is configured with a field equal to a predetermined array size,

When the database Open command is input, the array cursor is incremented by '1' until an empty field appears without a database connected to the array, and when an empty field is found, the value of the array cursor located in the corresponding field is set to the database ID. Database connection,

When the database close command is input, an array cursor is positioned in a field corresponding to a database ID, and the database connection is terminated after initializing the field where the array cursor is located by using the remaining operations using the database ID and the array size. Multiple database operating units in embedded database systems.
The method of claim 4, wherein

The database ID manager outputs a message indicating that the maximum number of database connections has been exceeded when the database Open command is input but there are no empty fields in the array, and returns a database connection failure. Multiple database operating units.
The method of claim 1,

A plurality of submodules are located between the buffer manager and the database to use a subordinate database by using database instructions that perform one or more logical functions.

The buffer manager implements by adding a parameter of the database ID to an application programming interface (API), selects one submodule among a plurality of submodules using the database ID, and selects a data page through the selected submodule. Multi-database operating device of the embedded database system comprising a submodule manager for reading.
The method of claim 6,

The connection manager uses an array data structure and sets the array size to the maximum number of databases that can be connected simultaneously.

The lower module manager selects a lower module to be used from among a plurality of lower modules by using a remaining operation using a database ID and an array size.
The method of claim 6,

The buffer manager caches a data page by displaying a database ID and a page ID in an internal buffer pool memory, and reads the data page from the buffer pool memory using the page ID and the database ID when re-requesting the data page. Multiple database operating device of the embedded database system, characterized in that coming.
Connection manager that handles database access to a plurality of databases loaded with data pages, a query processor that performs query processing operations including inserting, deleting, changing, and retrieving data into the data pages, and reading data pages from the database. In the multi-database operating method of the embedded database system including a buffer manager for caching and managing data pages,

The connection manager processing a database access through dynamic management of a database ID for generating or destroying a database ID for distinguishing a plurality of databases in order to use one buffer manager during run time of the database;

The query processor requesting a page request command to a buffer manager using a database ID and a page ID of a data page, and performing a query processing operation including inserting / deleting / modifying / searching data into the requested data page; And

The buffer manager implements by adding a parameter of a database ID, selects a corresponding database among a plurality of databases using the database ID according to the page request command, and reads a data page corresponding to a page ID from the selected database. And performing a role of caching and managing the data page in the buffer pool memory.
The method of claim 9,

The connection manager further includes a database ID manager that uses an array data structure, sets the array size to the maximum number of databases that can be simultaneously connected, and manages database IDs for distinguishing databases using array cursors. Multi-database operating method characterized in that the embedded database system.
The method of claim 10,

The database ID manager uses the array cursor as an unsigned short type, sets the range of the array cursor value to 0 to 65535, and includes a statement statement including close and open to control the array cursor. Multiple database operating method of the embedded database system, characterized in that using the.
The method of claim 11,

The connection manager may process a database access through dynamic management of a database ID for generating or destroying a database ID for distinguishing a plurality of databases in order to use one buffer manager, wherein the array is a field corresponding to a preset array size. If configured with

Checking whether a database is connected to a field where an array cursor is located when the database Open command is input, and incrementing the array cursor by '1' to search for an empty field when a database is connected to the corresponding field;

When there is an empty field in the array, the database cursor is assigned after assigning the value of the array cursor located in the field as the database ID, and when the empty field is not present in the array, the maximum number of database connections is exceeded. And outputting an informing message and returning a database connection failure.
The method of claim 11,

The connection manager may process a database access through dynamic management of a database ID for generating or destroying a database ID for distinguishing a plurality of databases in order to use one buffer manager, wherein the array is a field corresponding to a preset array size. If configured with

When the database close command is input, an array cursor is positioned in a field corresponding to a database ID, and the database connection is terminated after initializing the field where the array cursor is located by using the remaining operations using the database ID and the array size. How to operate multiple databases in an embedded database system.
The method of claim 9,

A plurality of submodules are located between the buffer manager and the plurality of databases to use a lower database by using database instructions that perform one or more logical functions.

The buffer manager implements by adding a parameter of the database ID to an application programming interface (API), selects one submodule among a plurality of submodules using the database ID, and selects a data page through the selected submodule. A method for operating multiple databases in an embedded database system, comprising a submodule manager for reading.
The method of claim 14,

When the connection manager uses an array data structure and sets the array size to the maximum number of databases that can be connected simultaneously.

The submodule manager selects a submodule to be used from among a plurality of submodules by using a remaining operation using a database ID and an array size.
The method of claim 14,

The buffer manager caches a data page by displaying a database ID and a page ID in an internal buffer pool memory, and reads the data page from the buffer pool memory using the page ID and the database ID when re-requesting the data page. Multiple database operation method of the embedded database system, characterized in that coming.