1.

5 CLASSIFICATION OF DATABASES
Databases can be classified in several ways, depending on the context.
that is being managed, or its usefulness.

According to the variability of the database

The first classification we will address depends on the way in which the
data is structured within the database. They could be none other than
the statics and dynamics.

Static databases
They are databases specifically designed for reading their data. Their
implementation in most cases is for storage and registration
historical data and develop studies that allow for understanding their behavior to
through time.
This is especially useful for companies that want to make projections.
statistics and guide the decision-making processes from the field
business.

Dynamic databases

As their name indicates, they are in constant motion, collecting

new data and deleting the oldest ones. Now then, with this data that
constantly striving to renew themselves will allow some company to
keep their query operations updated. Being useful for stores of
constant changes of products.

These are categorized into:

According to its content

Bibliographic database

They only contain their surrogate (representative) of the primary source, which allows
locate it. A typical record of a bibliographic database contains
information about the author
specific publication, etc. It may contain a summary or excerpt of the
original publication, but never the full text, because otherwise we would be in
presence of a full-text database.
 Full-text databases

Full-text databases are a very functional database that

allows us to search for specific terms, keywords, and all options of
a bibliographic database, with the great difference that in this database
We can consult the full text that is archived.
These databases are particularly useful for achieving objectives.
academics and scientific research.

Hierarchical databases

Hierarchical databases are, as their name suggests, a database of

data in which information is stored in a hierarchical structure
an order of importance. In this model, the data is organized in a
a figure that makes us remember an upside-down tree.
The hierarchical structure we achieve in trees is built
with segments that we know as nodes and branches.
The segments or nodes to build the tree can be in three forms or
categories:
Parent: it is a node from which descendants branch off. All parents are
located on the same level and have the same value of importance.
Child: it is a node that depends on the parent node. It can be said that it is a
derivation of the previous.
Root: it is the source of the data, as it has no parent node. It is located
At the top level of the tree. All the nodes branch out from it.
Network database
Network databases are quite similar to hierarchical ones; their difference
main in the composition of the node. In this model, the nodes can have
diverse fathers.
At the time, it represented a great advance over the hierarchical model already.
that its variation in the nodes offered a solution to the problems of
data redundancy.
Currently, this type of database is not used frequently since the
the difficulty of its modification and adaptation raises the degree of complexity of its
structuring it in a way that makes it very unattractive to programmers and users
finals.

Transactional Databases

These databases serve a very specific function. They are responsible for sending
and reception of data at high speed. Transactional databases
reality is uncommon for computer users who are not
related to the industrial and production field in complex lines.
A specific case in which we can have some contact with a database
transactional in a banking system that records exchange operations
of money between accounts.
From an account, data related to the due amount of money is issued immediately.
Destination account must generate the load data for the amount received.

Relational databases

Now we are on more familiar and somewhat more current ground. The foundations of
Relational data has dominated the field of databases for a long time.
The center of this model is the use of 'relationships' between data.
The predominant language in these databases is Structured Query.
Language known to all of you as SQL.

Its operation consists of entering all data into records, which

subsequently they are organized into tables.
Once the data is organized, the existing relationships can be established.
enter data easily and cross-check quickly to issue reports and
necessary analysis.

These databases are recommended when the data we are going to use
they have zero margin of error and do not require constant modifications.
It is based on entering data into records, which are then organized into tables.
These databases allow the elements to be related to each other in a very
simple and cross them quickly, establishing relationships between records in such a way
intuitive.

Multidimensional databases

These databases are designed for specific functions. There is no greater

difference between multidimensional databases and databases
relational. The point that separates them is only appreciable at a conceptual level. Already
that, in these, the fields or attributes of a table can be of two types.
They can represent dimensions within a data table or directly.
they can represent the metrics that are intended to be obtained.

Object-oriented databases

It's impossible not to get emotional when talking about the databases we use daily.
Object-oriented databases are among the most modern that we have.
In addition, we must not overlook its great capacity and power. In these, we do not
stores detailed information about the object, it is stored completely at
object.
The object is endowed with a set of its own characteristics to differentiate it from
objects that may be similar. The advantages of this model are obvious compared to
the ones described earlier. They allow for a greater amount of content and enable the
user to have more first-hand information.
 Documentary databases

Thedocumentary databasesthey are an interesting model of sets of

information used by documents such as storage structure and
data inquiry.
These documents are composed of multiple records and data.
They are built with NoSQL language which provides you with a large number of
technical and flexibility advantages.
This database model allows the handling of heavy volumes of
information in minimum time periods. Its diversity of functions and modules
adaptable to multiple query mechanisms has made it one of the
current preferred work models by programmers.

Deductive databases

A deductive database is a database that allows the possibility of making

deductions through inference. Its functionality depends on the
conditions and facts that are stored in the database. They are also
known as logical databases since their principles are
based on mathematical logic.
They arise as a response to the limitations that arise in databases.
relational when executing recursive queries and theorizing about the
indirect relationships that could be generated between the stored data.
This database uses a language called datalog that allows the
computer solve deductions to answer queries.

Among the main advantages that databases offer us

we have:
Use of logical rules to resolve queries.
It has stratified denials
Ability to obtain new information through that already stored in the database
of data through inference.
Use of algorithms that optimize queries.
Supports complex objects and sets.
We hope that with the information provided here, you will quench part of your thirst for
knowledge about databases.

1.6 DATABASE ARCHITECTURE

There are three important characteristics inherent to thesystemsof basesdata:
the separation between theprogramsof application and data, the management of multiple
views by users and the use of a catalog to store the
scheme of [Link] 1975, the ANSI-SPARC committee (American ...
National Standard Institute - Standards Planning and Requirements
Committee proposed aarchitectureof three levels for systems ofbases
of data,which is very useful when it comes to achieving these three characteristics.
The definition of asystemofinformationit is thedescriptiondetailed of the
system architecture. Database architectures have evolved
a lot since its beginnings, although the one considered standard today is the
described by the ANSI/X3/SPARC committee (Standard Planning and Requirements)
Committee of the American National Standards Institute on Computers and
Information Processing), dating back to the late seventies. This committee
he proposed a general architecture for DBMSs based on three levels or
schemes: the physical level, or machine level, the external level, or user level, and the level
conceptual. It also described the interactions between these three levels and all
the elements that make up each of them.
ANSI Architecture
The architecture of three-schema database systems was approved.
for the ANSI-SPARC (American National Standard Institute - Standards Planning
and Requirements Committee) in 1975 as assistance to achieve the separation
among application programs and data, the handling of multiple views by
part of the users and the use of a catalog to store the schema of the
database.

1. Internal level: it is the lowest level of abstraction, and defines how it

they store the data on physical media, as well as the access methods.

2. Conceptual level: it is the medium level of abstraction. It is about the

data representation carried out by the organization, which collects the views
partial requirements of different users and applications
possible. It is configured as a total organizational vision, and includes the definition of
data and the relationships between them.
3. External level: it is the level of highest abstraction. At this level correspond the
different partial views that the different users have of the database.
In a way, it is the part of the conceptual model that they have access to.

Theobjectivethe three-level architecture is to separate the programs

of application of the physical database.
The three-tier architecture is useful for explaining theconceptof independence
of data that we can define as the ability to modify the scheme in a
system level without having to modify the scheme of the immediate upper level.
Two types can be definedindependenceof data:
Logical independence: it is the ability to modify the conceptual schema
without having to alter external schemes or application programs. It
can you modify the conceptual schema to expand the database or to
reduce it. If, for example, the database is reduced by eliminating an entity,
The external schemes that do not refer to it should not be affected.
Physical independence: it is the ability to modify the internal scheme without
having to alter the conceptual schema (or the external ones). For example, it can
it is necessary to reorganize certain physical files in order to improve the
performance of theoperationsfor data query or update. Given
that physical independence refers only to the separation between the
applications and thestructuresstorage physics, it is easier to
achieve logical independence.

Therefore, the three-tier architecture can facilitate the acquisition of the

true data independence, both physical and logical. However, the two
levels of correspondence imply an extra expense during the execution of a
consultation or of aprogramwhich reduces theefficiencyof the DBMS. This is why
very few DBMS have implemented this complete architecture.
Computer networks allow for the separation of tasks in a client-server scheme.
servers, parallel processing within the computer allows for acceleration
some of the database tasks, as well as the possibility of executing more
transactions per second. Queries can be parallelized allowing for
that a query can be executed by more than one processor at the same time,
this feature has led to the study of parallel databases.