Home /
Design Database Designing Databases
The purpose of this report - to evaluate current problems and trends of computer aided design databases, as well, at least in part - the requirements of tomorrow day. The report could play another role: to specify a set of relevant requirements, which will serve as coordinates for the positioning of specific private methods and tools for designing databases (also partly - of their use and management), which appear in two special sections this conference.
Integrated database - a statement of ideas
It is widely known methods of designing of databases (DB) appeared in the development of increasingly sophisticated information systems (IS), which should have been consider the needs of more than one user, but large groups and teams. One such integrated database was created to solve many problems, each of which use only "their" part of the data, usually intersecting with parts used in other tasks. Therefore, the main methods of designing began a process of elimination of redundancy in the data. These methods are associated with other means of ensuring the logical integrity of data.
It was formulated fundamental requirement of separation of programs from the integrated data. This principle is designed to dispose of data as a resource enterprises is also important that the conservative nature of the data were separated from the application programs that can often be changed.
Another important problem of designing database was needed to ensure operational parameters, such as the amount of external memory or execution time of various operations. There are other requirements. For example, the information should not be lost not only because of equipment failures, but also because of user error. This is different from the situation where someone who solves a certain task, he is responsible for the safety of data for this task.
An understanding of the integrated database as a general information resource company. Stored data were similar to the big computer, which simultaneously used by multiple users with different goals and should be able to work all the time.
For an idea - a classic design methodology
Classical database design methodology - is a powerful and beautiful course with its philosophy, ways of perception of reality and modes of existence it. In this course there was his own applied mathematics, his notion of "Peace", "domain" (ABM) and their models. In the design database understood and integrated into a harmonious scheme of methods for performing such project phases:
collection of information on missile defense (needs analysis and description of the missile using the so-called "process" or up, "usage perspective" approach and "neprotsessnogo" or isp, "information structure perspective" approach);
choice of representation language called "semantic" model for the fixation of information about missile defense, their subsequent analysis and synthesis model database;
analysis of collected information about: classification, formalization and integration of the structural elements of the description of missile defense, formalization of both structural and procedural integrity constraints of elements in the future model of missile defense, the definition of dynamic object instances pts;
synthesis of the conceptual model database: designing an integrated conceptual database schema in the chosen language of the semantic modeling;
choice of a specific data model and database for implementing database;
designing logical database schema for the selected database (also known as "design implementation");
development of the physical structure of the database (the "physical" or "internal" circuit, it is the same - "layout"), including the placement of database nodes;
development of technology and procedures to create and populate the initial database;
development of technology and procedures for tracking database;
development of universal access programs to the database and corresponding user interfaces;
information support the development of specific data processing software: providing meta-information, data, test cases, etc.;
feedback from application developers and users of information systems (IS) on the completeness and effectiveness of the organization database;
testing database, its development and improvement (setting) of its structure.
There is every reason to call the methodology of classical: for these methods are developed comprehensive, holistic methodical system for the majority of methods proposed formal models, these models - or at least their final expressive possibilities - see the real application in practice of design. Only a list of basic data models and their authors produces very impressive, see their review, for example, in [Tsikridzis85].
We used the discipline called structural analysis at the project top-down approach. The structure associated with the use of hierarchical structures to detail data and functions, and relevant enough "hard" design procedures. The design scheme is called "cascade". It is well aligned with similar circuit design software - software.
For methodology - the workshop design tools DB
Designing an integrated focus on substantive, integrated, and, usually, larger database has become a complex task. The presence of a coherent methodology design allowed to take care of "shoemaker, the designer, and begin to sew his boots in the form of design automation systems database. This was facilitated availability of technological expertise in the organization and support of computer systems, software development and, on the other hand, the use of active integrated dictionary-reference data (dd / d, data dictionary / directory). Thus arose the system case (computer aided system engineering) - system for the structural design of the database and the associated IP-oriented data model, implemented in different DBMS. Most popular were case-system for relational databases with sql-data models, and dd / d was renamed in the repository projected case-IP.
In this way arose two main directions of development of case-systems and technology design: case-system for designing proper databases (or so Mr. upper-case), and integrated tools to design and database, and use them to develop applications. It is important to note that and upper-case, in general, have a lot of money to describe the functions of information processing (using a process approach to collecting and analyzing information on defense) and store these descriptions in the repository. This confirms the position of the strong coupling of the project database and project IP, based on this database. However, this relationship is not absolute, and the principle of separation of database programs persists.
Often, integration of functions leads to a strong merger case-system with one database on which the case-oriented application development tools programs. Such fusion has several manifestations, for example, case-supported repository means "native", but only one database, the generation of applied The programs are "native" development tools of the same database, but only them. For these integrated systems, case-mapping conceptual model database in a logical framework is often done well only for a predetermined database.
The latter fact is connected with another task that can be placed in the design of database: designing a portable database, which can be implemented platforms of different types of computers, operating systems, databases and even data models, and, if necessary, be transferred from one platform to another.
With that said, the classic studio designer database includes a set of classical methods of structural design, a set of relevant modeling tools, implementation, load and maintain the database, as well as the "cascade" organizational chart of the implementation of these works on the principle of "top down.
Especially - the time characteristics and transaction
Ensuring the performance of database - still a challenge despite the increase in specific power of computers and reduce unit costs memory. This definition of the temporal characteristics of working with databases and the preservation of these characteristics in the operation database refers to the most difficult project problems. The design phase to determine a rational physical database schema of the ways to determine the temporal characteristics should be as follows:
able to compare the time-frame options for implementing different versions of the database schema, to a certain database;
able to compare the options for implementing a database schema in different databases;
able to compare the implementation of a database schema on different hardware servers, DB;
to predict the time parameters of the various software applications and utilities, utilities.
Objective comparison of temporal parameters of different databases is considered an independent. However, it often must be addressed as part of the design problem of choosing DBMS for the projected database and in the process of designing.
The notion of transaction was introduced to determine the complete set of actions on the database, which converts a database from one integrated in a logical sense state to another. At its base were built, primarily, mechanisms correctly update and restore the database. But then on this basis are based and other mechanisms and methods.
Interim evaluation of the most popular DBMS tests recently given in the form of a number of transactions in certain types of standardized unit of time. Distributed processing is based on the transaction monitor.
It will be necessary to detect the limits of such division of work at a fairly small portion. Here we note a very important effect: the practice of targeting on "transactional" approach is closely connected with the classical methodology of designing the database, which was developed mainly as a design methodology So-called "operational" databases, ie databases, which should fix some of the transactions and to keep the current model of the actual state object or missile defense.
The assessment of the state
It is possible that the reader the impression that we already possess modern methodology or at least close to it, that, unfortunately, not so, and maybe we will never achieve anything like it. Always difficult to describe the methodology at a conceptual level, very difficult to apply it into practice. The stumbling block - the difficulty of penetration into the essence of the subject area (eg, difficulty understanding the mechanism of the organization) and adapt it to new, possibly better, the conditions of operation.
Similar problems are characteristic of the database as a whole. The database system should be integral to the system management of the organization - this is a guarantee of its successful application. However, the process of its implementation is linked to certain changes in the organization and activities of its members, and we will always deal with the natural inertia of the people when it comes to perceptions of change ....
It is important that the funds were adequate to the needs of database users. Since different users may require different data models, languages data and schema, it is desirable that the database supports a variety of means, and the user can choose the one most suitable. ...
You can certainly call into question the value of such research. Indeed, no matter how bad a programming language, it ultimately still can learn. Likewise, DBMSs and tools can be mastered over a period of time. But the problem is not in development funds, and in the effectiveness of their use. The machine should be a servant of man, but not vice versa!
Above the bold highlighted quote from [Tsikritzis85]. Since then, the DBMS, database design methods and tools have considerably added to the possibilities. But the rest of the world is also not standing still, will be difficult challenges facing developers of IP and database tasks. And we must admit: the language Tsikritzisa Lohovski and not obsolete.
What and how of classical design methods used in practice so far
Applied in practice:
DBMS, supporting a relational data model in 1971 with some extensions (see, for example, [Dadli96]);
Hierarchical "cascade" scheme of the structural design database at the approach of "top-down";
case-system for the structural design of databases, IP in general, or, moreover, application of IP. The most commonly used: options er-model data tabular relational model in 1971, extended by a further set of descriptions of integrity constraints (referential integrity, business rules) for the analysis of a "process" a source of information often provides a model for data streams or sadt, possibly enhanced by additional relations management (these links should not be confused with the allocated flow conditions for carrying out functions in the notation idef0);
Utilities dynamic database administration, providing the following functions:
tracking dynamics of the operation of databases: indicators are available at any time against the background of the application, these indicators (statistics) can be used to support the dynamic creation of optimal access paths to data
backup database, as well as keeping copies of a hot standby database on the background of the application, recovery and rollback segments and the complete database
possible dynamic database reorganization (relocation of the database and the separate pieces of physical, logical and physical restructuring of the data), however, these opportunities are limited.
Accounting for custom reporting requirements in a greater range than before. Accounting for the specific representations often become transformed the provisions of the desirability of having different external data models to the situation of the availability of a large number of custom tools that have different interfaces and, in practice, different external data model.
What is lost
However, much of the classical heritage in practice is not used or used badly. Primarily, this relates to the use of formalized methods and models, unless they are included in your data model directly, but must be used by designers to obtain and verify High-quality design solutions, for example:
the whole procedure of normalization of high degrees and minimize the set of relations is not carried out or are rarely, if the examination of compliance testing even 3NF or BKNF provided for in the case-tools, this feature is rarely used in practice because of its bulkiness and high skill requirements designer uses the case;
optimization of placement database to external storage devices is carried out "by eye", issued today tests the time parameters are not adapted to assistance in this task design;
just "by eye" is optimized placement sites, distributed database for database.
Much less attention has recently been given, and tools, automation of physical database design, including mathematical and natural modeling characteristics of the database, including - on the deployment of nodes distributed IP. Optimizing placement of nodes of a distributed database DB is not supported by common case-means. Separate tools and work, including domestic research, do not "weather" in Workshop of design, and do not support a living school of this trend.
This is, in our opinion, several reasons:
high demands on the skills of designers in the field of theoretical foundations of classical design database;
awkwardness of the methods used in the "cascade" scheme, in terms of practical impossibility to ensure the sustainability of large integrated solutions in the world with constantly changing requirements of the IP;
relative ease of implementation of the reorganization of the logical and physical structure of the database in relational databases (however, and this is specified at the end of the report, in modern conditions, this approach becomes one of the pitfalls for the designer).
Limitations of classical methods
Classical models and methods relied on the organization of storage and processing of structured data in detail, which corresponds to the notion of "attribute" as properties of the object represented by the atomic data element. In consequence, for example, full text databases at once stood out in a special type of database. For their design, there was a separate course - or the IRS Information Retrieval System.
After years and decades after the announcement of its classic models and concepts of classical explicitly describe their limitations. For example, in [codd79] was stated that "when discussing the semantics of data modeling focuses on the structural aspects at the expense of aspects of processing. structure without appropriate operations or implied methods similar to the anatomy in the absence of Physiology.
Even after 14 years E. Codd et al in [codd93] fixed: "... possession of a large corporate database is of little value if the end users have the ability to easily synthesize the information from these stocks (warehouses) of data. Too often we have just such circumstances. "
Finally, the time has come when designing databases (and IP in general) on the classical rules of completeness and integrity often become virtually meaningless. Wesley P. Melling (garthner group) pointed out in [Melling95]: "By 1990 almost all aspects of" standard operating procedures "with IT (Information Technology - Approx. EZ) were challenged, and computing architectures burst out of control. ... Programming standards erode, and the concept of nonredundant, consistent, high quality data was suitable except that for piles of rubbish. "
The reasons for the emergence of new requirements
The phenomena of global computer communications and pervasive personal computing (with a fall in unit cost of computer equipment) led to many new features and database design. The list of types of stored and processed data has grown to the extent possible, defined the most common normative meaning of "given". In the corporate database includes not only the raw elements and full-text fragments, but also Database with geo-multimedia database, and it is not an exhaustive list.
Moreover, new opportunities for IT - along with a number of purely economic reasons - have led to increased market opportunities and demands of consumers, as consequence - a sharp increase in competition in various industries and services. The answer was the announcement of the imperative of business reengineering: from bpr M. Hammer ([hammer93]) to the construction kiberkorporatsy by J. Martin ([Martin95]). These approaches require the implementation of radical changes in the organization of the main activities of enterprises. In particular:
a sharp decline in time spent, number of employees and other costs of execution of production functions;
Globalization of business: Work with customers and partners anywhere in the world, as well as work with the client in a mode 24 * 365;
reliance on the growth of mobility of staff, provision of opportunities for all employees can obtain the final result;
Work on the future needs of the client, the rapid progress of new technologies.
If IT was one of the shocks to this new situation, they were called, and in order to ensure the success and the possibility of the planned reconstruction. There are new requirements for the architecture of corporate IP, as a result - new requirements for the corporate database.
Also, as the IP itself can not be separated from its users, and the new design should be considered as an integration of three components: requirements of business re-engineering, human factors and methods of new IT. The real unification of these three components, each of which acquired in the 90 years of qualitatively new content, created the beginning of what could be called the New System design - NSP.
What you need from the database to respond to new demands
We will show the new requirements to corporate databases on the example of two aspects of the creation of new corporate IP (more than two dozen types of work components based NSP - See [Zinder96]):
Maximizing opportunities for each employee, then there is support for the implementation of all business functions, thus the employee, who receives final result. From the IP database and the database that requires:
providing means of access to all necessary data using a distributed database of data replication, event management data and processes transaction processing;
the use of architecture and software for data storage, means of online analytical processing (olap), the use of rapid application development (rad) to create a "head of IP" (eis), decision support (dss) on the data warehouse, olap and rad / eis;
use of dss-based analysis of database of precedents, as well as methods of inference, neural networks and neuro-computers, etc.;
proposal to a single user interface to work with different components of data and applications that use this interface tools that enhance easy to find information and treatment to specific application functions, such as functions geoinformsistem, hypertext, natural language, speech input.
Development of the concept and structure of the corporate database to the new IP, the implementation of the database structure, suggesting the removal of (significant reduction) restrictions its development, including, by changing the functions or functional components of information processing:
application of methods of designing meaningful component database for operational database and historical database for data warehousing, document archives, geo-information and other data;
Develop procedures component changes the corporate database when you change business processes, activities, used applications and geographic enterprise;
constant updating of the conceptual model of the enterprise to integrate new concepts arising from a change applied to the functional component similar and when you change activities, enterprises, and building on this basis, new interfaces between the components of the IP;
dynamic administration of fragments distributed enterprise database when you change the frequency of their use, with modification of their structure and change their placement.
Some new technical requirements for the databases can be obtained from the analysis in [Melling95].
Entered into practice new tools studio designer
Language sql, a former 80 th year, only one of the languages of the relational model, has become a real standard, not only relational data model, and industrial DBMS. (At the same time, it is - an example of the acquisition, which quickly can become burdensome.)
In the actual production of the most common organizational and production of IP in most cases or in most of the volume of work was replaced development tools with sql in the switch 3gl-language or language 4gl procedure type on the languages and tools 4gl with window interface based management through the menu and using the elements of the concept of object-oriented programming (and the preservation of opportunities out on the sql and procedural language).
There were virtually running a de facto standard interoperable work with data in the first place - a standard odbc.
Multiplatform become the norm, multiprotocol communications for distributed database implemented on the basis of standards and interoperable monitors transactions supported "internationalization" at least in part on the configuration of the table of national character sets of data.
We went into practice the new structures and data types, new operations on the data: the raw elements, full text databases and their processing, GIS data, multimedia Database, hypertext distributed database, distributed processing and handling, together with the object delivered by the input of the IP. In practice, there are real steps integration of these structures and operations.
Changing the approach to the choice of DBMS in the first place - for the design of corporate databases, maintenance and development of which is planned for at least a few years. Increasingly used by economic reasons and criteria for selecting databases (see [Zinder95a]).
Object orientation in the design database is not considered as already exists in practice a new instrument Workshop. (It is not intended object-oriented programming.) currently it seems reasonable to attribute such a design is still to the directions of research.
New approaches to organization design database
Since the new requirements in a large, if not determining the degree associated with increasing rates of change in the requirements for integrated circuits, new approaches in the methods of designing inextricably linked with the new organization design.
Cascade schemes of designing software for integrated circuits have long been converted into a cyclic form. Thus, the organization continues to develop ibm corp. (see [Foks85]) provides continuous controlled development of the software system in the form of transfer of operation of its new versions.
We have different cyclical and spiral circuits are considered as a means of taking advantage of rapid prototyping approaches with the exception of IP their deficiencies (unmanageable) through the use of classical structural methods at every turn of the helix.
However, such cyclic schemes retain many structural shortcomings of the old methods. For conditions of NSP important deficiencies are:
complexity of changes in existing components;
limit the ability of component design uses a complete set and perekomplektatsii various ready-made components.
There are other, for example, the complexity of the project documentation. All this fully applies to the design database.
In terms of component design an organizational chart design database should look like a parallel circuit of the spiral design components DB and their complexation of necessity.
You can often find the claim that object-oriented design and programming solve the problems caused by the structural approach and the remaining using cyclic circuits. However, using technologies such semantic interoperability issues, especially - when the component design, may in fact even more complicated due to encapsulate descriptions and pay less attention to the discipline of documenting data. Seems which to draw conclusions about the limits of applicability of these approaches is premature.
From the new requirements for the types and sources of information - to the new architectural principles DB
The most important task of designing the architecture of corporate database is to work with a variety of types and sources of information. Under with the actual practice of sources and information consumers are not only units of the company, headquartered holding company or ministry staff, but companies in other industries (potential suppliers and consumers, government regulators and others). The principle of globalization of business dictates: sources and consumers of information will be located in any geographical point where it will be necessary.
Hence the strategic architecture database and IP in general solutions. Combining the requirements for the dynamics and diversity of types of information flows processed in IP, with the increase of their volumes, and the requirements for the diversity of processing methods can give the following generalized characteristics of technologies that form database architecture consisting of integrated circuits:
component technology design and perekomplektatsii object-oriented operational database, allowing the user through common, in including, for the data warehouse, interfaces;
Extended data warehouse technologies, integrating historical formatted data, archives text documents, audio and video archive, as well as cartographic data, and includes tools for online analytical processing of data, necessary types of "friendly" interfaces, such as: hypertext, GeoInformSistem etc.;
open database to incorporate and obtain information from it using the principles of the global information highway;
Open-system architecture, advanced methods and means of forming a component: the upper level is open component design Database and the free exchange of any external sources of information systems at the lower level - the technological openness of databases, standards-based portability interoperability, scalability, and others.
Advanced technology makes the Integral warehouse on a new basis to question the development of an integrated set of user interfaces, which created a natural environment for working with information and functions regardless of which class of data storage developer has to attributed this to (user) information.
New approaches in the methods of designing database
In response to the new requirements may be considered recommendations for new methods and tools for designing databases. (Of course, the assumption that all new have previously someone had discovered the old.)
On the exclusion of redundancy in the data
Is saved as a reasonable requirement of a single data entry in the database for the solution of various problems and protect against a conflict (violation of the logical integrity) for updating the stored data. However, in the global information space and component design context of requirements should be revised. Undoubtedly, the database operating rationally plan the "island" normalized and, in the classical sense, nonredundant clusters of relationships or objects. These "islands" often and will be the long-known-focused database.
However, the association of historical data storage, databases, GIS systems, archives text documents, information flows received by the Information Highway and others in the general formulation of designing a corporate database leads to a denial of the universal principle of exclusion and vseobyazatelnogo redundancy: design corporate database at the logical schema and the conceptual level is not supported as a global criterion on the requirement and procedures for removing redundancy in the data.
The new situation, including the substantial and previously unregulated flow of information from the information highway in the corporate database will require the development or increasing the capacity of "identification procedures" copies of information structures, ie determining that these instances describe one and the same view of the real world.
The problem of conservation issues
The very nature of the discipline of design, provided the cascade scheme, the methods of structural design and hierarchical approaches and structures encourages designers to fix now rather strictly defined model of missile defense. Should be changed case-technology design databases such way that no conservation of the existing problems of enterprises in hard, "tselnointegrirovannyh" databases. This may require changes not only technology but also design tools.
Estimated approaches:
to capture descriptions of attributes, entities, relations, functions, etc. with any degree of incompleteness, the ability to produce descriptions at nedetalizirovannyh, substantively related sets of information structures ( "clusters of entities");
design or reconstruction of models of IP components and databases, their integration into the overall conceptual space;
Designing an ordered sequence of states of the corporate database as a sequence of sets exploited subject database, including: inherited database structurally predefined database "purchase" functional components, designed specifically for the enterprise database, and DB last two categories are gradually replacing the inherited and then in parallel, replace each other in the development of IP;
open repository case-system, the dictionary database and the system 4gl, allowing overbuild metaobjects and mechanisms required Thesaurus and deep semantic relationships between elements, as well as to produce a two-way exchange of meta-information with other systems and 4gl case, connect the different models components into one, using and maintaining all necessary semantic relations.
Component openness and semantic interoperability
Component approach in the development of integrated circuits requires the design of the component database. Replacement of some functional components of the IP in a similar, but engineered another developer, will require structural change some of the corporate database. Such replacement shall be maintained as an ongoing process re-engineering DB. When replacing the components of database interfaces with its existing applications and their users should receive exactly the same in regard to the semantic information that earlier.
Real component design database may be based on the formation and use of common kompleksiruemyh component of the conceptual model and maintenance correspondences between the models of the component databases (and associated applications) and the overall conceptual model. In general, the requirements for formalisms such models described previously (see [zinder90]). Recently developed software implementation of complete formal systems, usually based on the object approach, which may approach the instrumental to this task (see eg [Kalinichenko93]).
Development of conceptual models and SKK
The need for the use of common conceptual models makes re-examine the problem of designing database of what is called NSI (normative-reference information) and CCM (the system of classification and coding).
Until now common view that the JCC - a means of reducing the submission of information in an integrated database. In fact, the absence of CCM or the use properly constructed CCM leads to the incompatibility of semantic information stored in different databases or even in a single database. In these conditions does not lead to goals using the most advanced technological modes of interoperability. Thus, it is advisable to use the work on designing DB with NSI and design of the CCM as the beginning and the basis for a conceptual space missile defense, to build a conceptual model of the enterprise.
Conceptual models and subsequent design work
In later stages of designing the actual database conceptual model continues to be used for various purposes, such as:
different subject for a set of information models with the release of a common data entities;
development of functional models of various types;
development of semantically rich user support tools, etc.
Technological breakthroughs
In the IP architecture of the new database will likely but not the only component of integrating software (including - intermediate or middleware). Monitors transactions and processes, means of semantic modeling and the use of conceptual models, DBMS-independent development tools and execution of applications - Other classes of software components that achieve this goal.
It is recommended to preserve the independence of the database through the use of tools and standards covering various DBMS. Failure of connection with one DBMS, open case-management repository, the opportunity to develop metamodels supported in the repository, and applying to them the design procedures - this is only the minimum requirements for the methods and tools.
It is advisable to focus on case-driven systems, the possibility of parallel design component independent developers (including including - without the use of this case-system) with subsequent integration of meta-information.
Application Development must meet the requirements of mobile applications and, simultaneously, work in a heterogeneous distributed database environment.
Problems volumes, time characteristics and physical design
Distribution Database class vldb requires greater use of methods of designing effective physical data schemas. It is impossible to build such DB counting on continued restructuring by the census in the new physical structures. This is so for the operating mode DB oltp, the more so for terabyte database oriented olap. Ease of implementation procedures reorganizations this method may become a "trap" for the designer, especially - in the early stages of writing a DB in action when its census is still possible due to incomplete volume.
It is appropriate at the level of new technologies (the use of multidimensional structures, indexes, bit-maps, etc.) to return to the methods of forecasting performance characteristics of the database, which would allow to plan the physical stability of the scheme at least while the economic opportunities will not expand external memory at various levels for other approaches.
Large increases in the database will be accompanied by increasing demands for their reliability. Means and methods for the DB design due to continuously implemented process re-engineering of database will be directly adjoin controls database. Thus, to ensure fault tolerance of data requires a command management and synchronization of geographically dispersed shadow and standby databases.
The problem limits the applicability of the two main methods of designing
In the course of research and practical design should be defined the limits of applicability of two concepts: design database as an object, realized separated from application programs and object-oriented design, in which the object encapsulates both data and methods for their treatment.
CONCLUSION
Creating a corporate database in the new system design - an activity that uses many techniques of classical design, but requires other organizations, and many additional techniques, as well as new, which would replace some of those that were developed more than 10 years ago.
The discipline of designing databases in the new environment is still missing. Nevertheless, its beginning can be seen, its elements are working in real projects.
In accordance with the principle of conservation of immunity to the computer revolution (see [Zinder95b]), classical methods of designing databases should continue to be used, but only those in areas where they are really useful. Design methods discussed in specific projects of corporate IP and the database, and relevant instruments should be checked for their ability to provide functions in accordance with the requirements of the new system design.