RU2606315C1 - Method of processing user requests by distributed information system - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: invention relates to database management systems (DMS). Method is proposed, in which in distributed information system summary database is made storing all changes of remote databases, wherein said changes are recorded with their marks at level of access and are stored in inner table with their identifiers; as soon as summary database obtains new record, message on changes in remote database of distributed information system is generated and transmitted to users operating in real time; service information is recorded in log file on received request; service information is compared with other requests of records log file; if there is complete coincidence of service information of coming request with already available in log file of service information, request is characterized as request for update, and in this case, data for answer to DMS core is read from summary database.

EFFECT: technical result consists in reduced time for information exchange by treatment in DMS by means of summary database in remote database update requests on distributed information system and due to possibility to transmit data of changes to users in real time.

1 cl, 2 dwg

Description

The invention relates to computer technology and can be used to create and improve database management systems (DBMS) and distributed information systems.

There is a method of updating information in databases of control objects of an automated control system, which allows to reduce the time of information exchange and provides the ability to transfer changes and additions that have occurred in the databases to users in real time (patent for invention of the Russian Federation No. 2409841, 04/20/2009).

The disadvantage of this method is the significant time resources for exchanging information due to the large number of operations necessary to process user requests and sending service information via communication channels to update information in databases, and the method does not provide for access to confidential data authorized users and denial of access by unauthorized.

The closest in technical essence and performed functions analogue (prototype) to the claimed one is “A method for providing access to data in the Linter-VS database management system” (invention of the Russian Federation No. 2130643, 10.21.1998), which consists in remembering in advance the individual identifier signal and the rank access signal to the database assigned to each database user, send the request signal with its identifier and descriptions of the requested data to the DBMS kernel, read the user access rank by its identifier, soda that is, requesting information from the database, the signals corresponding to the labels of these data characterizing the level of access to sections and subsections, data, as well as actions that users can perform on this data, are read into the DBMS kernel according to the descriptions of the requested data from the database compare them with the rank of the user’s admission, and if the user’s admission rank is not lower than the access level to the database section containing the requested data, open a communication channel with the database, generate link signals to the requested data After updating their descriptions in the request, the request signal is divided into separate actions, during which they are registered in the DBMS kernel, the response to the request is generated in the DBMS kernel using the results of actions performed on the links to the requested data, the data is read into the DBMS kernel, and from there - to the user.

The disadvantages of the prototype method are: significant time resources for the exchange of information due to the fact that the possible distributed nature of the information system is not taken into account and requests for searching and updating data are processed using the same algorithm, as well as due to the lack of the ability to transmit to users changes in the databases of the distributed information system in real time.

The objective of the invention is to develop a method for processing user requests of a distributed information system, which allows to expand the capabilities of the prototype method, as well as to reduce the time resources for information exchange through processing in a DBMS using a consolidated database of update requests in remote databases of a distributed information system and due to Providing the ability to transmit changes to databases in real time to users.

This problem is solved by storing an individual identifier signal and an admission rank signal to a database previously assigned to each database user, generating a user request signal containing an individual identifier signal of this user and description signals of the requested data, transmitting the generated user request signal to the core of the control system databases, read the signal of the rank of admission to the database by the signal of an individual identifier, read from the database to the core of the system Database management by the signals of the description of the requested data contained in the signal of the user’s request signals the signals corresponding to the labels of these data, characterizing the signals of the access level to the sections, subsections of the database, to the data, as well as to the actions that the user can perform on them, compare the signals the access level with a signal of the rank of admission to the database, if the signal of the rank of admission to the database is not lower than the signal of the access level to the database section containing the requested data, allow the issuance of data signals via a communication channel from a database, according to the invention, the method is supplemented by the following actions: in a distributed information system, a consolidated database is created that stores all changes to remote databases, while these changes are recorded with their labels by access level and stored in an internal table with their identifiers; as soon as a new record appears in the composite database, for users working in real time mode, they form and transmit a message about changes in the remote databases of the distributed information system; write service information in a log file about the request; comparing the service information of the incoming request with the rest of the log file entries before comparing the access level signal with the access rank signal to the database in the core of the database management system; if there is a complete coincidence of the service information of the incoming request with the service information already in the log file, then the request is characterized as an update request, in which case the signals corresponding to the data labels are read from the composite database, the data for response to the database management system kernel is read from consolidated database.

Thus, the time resources for information exchange are reduced due to the possibility of transmitting changes in the databases of the distributed information system in real time and the lack of the need to look for and read updates in the remote databases of the distributed information system, spending time transmitting information through communication channels.

The analysis of the prior art made it possible to establish that there are no analogues that are characterized by a set of features identical to all the features of the claimed method for processing user requests of a distributed information system. Therefore, the claimed invention meets the condition of patentability "novelty."

Search results for known solutions in this and related fields of technology in order to identify features that match the distinctive features of the prototype of the claimed method showed that they do not follow explicitly from the prior art. The prior art also did not reveal the popularity of the impact provided by the essential features of the claimed invention transformations on the solution of the problem of the invention. Therefore, the claimed invention meets the condition of patentability "inventive step".

"Industrial applicability" of the method is due to the presence of a developed modern elemental base, on the basis of which devices can be made that implement this method with the achievement of the destination specified in the invention.

The claimed method is illustrated by drawings, which show:

FIG. 1 is a block diagram of the processing of user requests of a distributed information system;

FIG. 2 is a graph for estimating the length of time required to process requests from end users of a distributed information system.

The claimed method is illustrated by the structural diagram of the processing of user requests of a distributed information system (Fig. 1).

In the invention, users appear in the system after the process of assigning an individual identifier to this user (hereinafter referred to as the user identifier).

All requests from users who form queries to the database of a distributed information system (block 2) are transmitted to the DBMS (block 3). When accessing the database, the user generates a request signal, which will contain information about the nature of the request: a search or update request, as well as information about the user's desire to receive all changes in the databases of interest to him in real time.

Throughout the work of a distributed information system, it is in it that a consolidated database is formed (block 4), which stores all changes to remote databases. As soon as a new record appears in the composite database, for users working in real time, a message is generated and transmitted about changes in the databases of the distributed information system (block 4).

Initial processing of the request takes place in the DBMS (block 5): the generated request signal is transmitted to the core of the database management system, the signal of the rank of admission to the database is read by its individual identifier signal contained in the user request signal. The DBMS writes service information about the request received in its log file (block 6). Under service information in the invention is understood the type of operation performed by the user (insert, select, update, delete), the name of the table with the requested data, the column number and values in it.

The DBMS, which is part of a distributed information system, compares the entry made in the log file about the incoming request with the rest of the entries in the same file (block 7). If there is a complete coincidence of the service information of the received request with the information already available in the DBMS log file, the request is characterized as an update request and the DBMS refers to the consolidated database (block 8), which stores all the changes that occur in the databases of the distributed information system and is expanded in read-only memory on the same computer as the DBMS or to remote databases (block 9), it finds the signals of the labels of these data contained in the received request signal description signals and reads them for further comparison nenia. These label signals characterize the access level signals to the corresponding sections and subsections of the database where the requested data is stored, the access level signals to the data itself and the access level signals for actions that can be performed on this data. All these access level signals are recorded in the DBMS kernel, in its memory for the entire time the incoming request is processed. Then the DBMS kernel compares the admission rank of the user who sent the request with the access level signal to that section of the database where the requested data is stored (block 10). If this signal of the rank of admission to the database is at least not lower than the signal of the access level to the partitions with the requested data, the DBMS core sends the corresponding command to the multiplexer or other means providing the opening of a communication channel for reading data from the database (block 13). This creates the opportunity for subsequent data transfer from the database to the database engine.

After all the required actions (groups of actions) are performed on the signals of links to the requested data, based on the results of their execution, the DBMS core generates a response signal to the incoming request signal in its working memory (block 14 and transmits the generated response to the system user (block 15).

If, after comparing the user’s admission rank with the data labels, the user’s admission level is lower, then the user is generated (block 11) and transmitted as a response to his request (block 12).

In this case, the remote databases, when changes are recorded in them, form a package with the necessary service information about these changes and transfer them through the DBMS to the consolidated database, where they are recorded with their labels by access level and stored in the internal table with their identifiers. Thus, if a user of the system sent update requests with a real-time processing request, then after appropriate checks of his request signal, he will also be notified through the DBMS of the changes that have occurred in the remote databases.

In order to determine the nature of the request in the DBMS, a log file is kept to record the progress of transactions in the databases. The journal contains records of transactions and its functions are not only to restore, but also collect information about current performance, accumulate control information, etc. (Databases. Design, implementation and maintenance. Theory and practice. T. Connolly, K. Begg, A. Strachan. Second edition. Williams Publishing House. Moscow. St. Petersburg. Kiev. 2000. Page 627) .

In the proposed method, all service information and actions on request will be recorded in the log. By adding a new entry to the journal, it will be compared with existing entries and, if the entries match, the request will be defined as an update request.

Formation and sending through the DBMS of information packages to the consolidated database and, subsequently, after recording the corresponding information in the consolidated database, the message about the changes is sent to users using triggers and the “notify” command.

Triggers determine the actions that the database must take in the event of certain events related to the database. The execution of triggers is transparent to the user. Triggers are executed by the database from the moment when specific types of data manipulation commands are executed on its specific tables. (Oracle Database 10g. Complete Reference. Kevin Looney. Laurie Publishing House, 2004, 570-573). For a composite database, this is an insert; for other databases of a distributed information system, this is an update command.

It is proposed to organize the composite database block using modern database technologies. Structures such as tables and indexes will be created in the master database. The master database instance will include memory called the System Global Area (SGA) and background processes that interact between the SGA and the database files on the disks. The main structure of the master database will be tables. In the composite database, it is proposed to use index-organized tables, the data of which is stored in the index structure, which allows them to be sorted inside the structure (Oracle Database 10g. Complete reference. Kevin Looney. Publishing house “Lori”, 2004, p. 5).

The signals corresponding to the data labels and the databases themselves are stored in the main computer memory on which they are deployed. Therefore, the reading of the signals corresponding to the data labels and the data itself from the database is organized as reading from the main memory of the computer. The main memory is formed by random access memory devices. Such storage devices are formed as an array of cells, and “random access” means that accessing any cell takes the same time and can be done in random order. Each cell contains a fixed number of storage elements and has a unique address that allows you to distinguish between cells when accessing them to perform write and read operations (B.Ya. Tsilker, SA Orlov. Organization of computers and systems. Peter. 2004. p. 203 )

In the same way, information is recorded in the DBMS log file, which is also stored in the main computer memory.

Bit indexes are supported to configure access to tables in the master database. They are used in situations where data is downloaded in batch mode. (Oracle Database 10g. Complete Reference. Kevin Looney. Laurie Publishing House, 2004, p. 7).

The claimed method for processing user requests of a distributed information system provides the ability to process update requests in real time, reduces the processing time for update requests, thereby reducing the total time of information exchange in the system.

Since the assessment of temporary resources for processing user requests and temporary resources for transferring changes and additions made to remote databases (information exchange time) are directly related to the time required to process the number of all individual operations that are necessary to complete the request, achieving a technical result by reduction of temporary resources for information exchange (transfer of changes and additions to general information databases to end users of distributed information specialized purpose systems) can be represented in the form of a graph (Fig. 2) comparing the results of the information exchange time of the claimed and known data processing method depending on the percentage of update requests that require processing, according to the claimed method, significantly less time to search for the necessary information, accordingly, less time for their processing, from the total share of all requests from the client block for their further processing.

The reliability of theoretical proposals was checked using simulation models of a prototype system and a system that implements the claimed method of processing user requests for a distributed information system for specialized purposes.

The simulation was carried out on the basis of the theory of queuing based on the GPSS modeling language. The experiment was conducted in the GPSS World software environment.

The experimental conditions were the following initial data:

1) the number of nodes with databases n = 300;

2) loading the DBMS by processing update requests A = 0.3, where A = λ⋅τ, while λ is the intensity of receipt of update requests, and τ is the average time for processing update requests;

3) loading the DBMS by the processing of search queries A = 0.015, where A = λ ,τ, while λ is the intensity of receipt of search queries, and τ is the average processing time of search queries;

4) the ratio between the time spent on processing the search query and update request γ = 3.

A comparison of the experimental results shown in FIG. 2 shows that the application of the proposed method gives a significant gain in reducing the time resources for information exchange (depending on the share of update requests from the total number of all requests from users of the request system) compared to the prototype method.

Claims (1)

A method of providing access to data, which consists in storing an individual identifier signal and an admission rank signal to a database previously assigned to each database user, generating a user request signal containing an individual identifier signal of this user and description signals for the requested data, transmitting the generated request signal the user to the core of the database management system, read the signal of the rank of admission to the database by the signal of an individual identifier, read from the database to the core of the database management system according to the signals of the description of the requested data contained in the user request signal, the signals corresponding to the label signals of these data characterizing the signals of the access level to sections, subsections of the database, to data, as well as to actions that the user can perform on them, compare the access level signals with the access rank signal to the database, if the access rank signal to the database is not lower than the access level signal to the database section containing the requested data allows the issuance of data signals over the communication channel from the database, characterized in that a distributed database is formed in the distributed information system storing all changes to the remote databases, while these changes are recorded with their labels by access level and stored in an internal table with their identifiers; as soon as a new record appears in the composite database, for users working in real time mode, they form and transmit a message about changes in the remote databases of the distributed information system; write service information in a log file about the request; comparing the service information of the incoming request with the rest of the log file entries before comparing the access level signal with the access rank signal to the database in the core of the database management system; if there is a complete coincidence of the service information of the incoming request with the service information already in the log file, then the request is characterized as an update request, in which case the signals corresponding to the data labels are read from the composite database, the data for response to the database management system kernel is read from consolidated database.

Images