por The compilation of information and transmission of knowledge has always been a central concern in different human civilizations. It is what motivated the emergence of writing or any symbolic representation that allows us to accumulate knowledge and expand our understanding of what surrounds us. This is also the origin of the topic that we are going to address in this post: what is a database? Let’s get to the point.
What is a database?
Let’s start with the question that gives the title to this post, what is a database?
This is any set of data that shares one or more characteristics in common and that is stored for later consultation and exploitation. In this way, a library is a database.
However, when we talk about databases today, We refer to those that are in digital format, since they now represent the majority. In this context and according to the definition of the Pan-Hispanic Dictionary of Legal Spanish, the concept refers to a “computer memory in which data arranged so that they are individually accessible by electronic or other means can be integrated.”
In any case, a database can contain various types of data, whether quantitative or qualitative, numerical, alphabetical or algorithmic.
This data can be structured -when its characteristics are already defined and are classified in tables such as dates-, unstructured -when they do not have a common default configuration such as text documents or multimedia files-, or semistructured – that have both predefined elements and specific markers, such as HTML and XML documents.
At the same time, a database is not the same as a computer application for storing and displaying information, such as spreadsheets. What differentiates the first from the second is their level of complexity and the large amounts of data they are capable of managing; all this, without leaving aside efficiency in user accessibility.
Therefore, companies and institutions They need database managers to facilitate access and use of information. These are the Database Management Systems (DBMS).
Database Management Systems (DBMS)
Once we understand what a database is, let’s see how the systems that manage them are articulated. A database management system It is a set of tools that allows you to store, define, consult, update, process, analyze, make backup copies…, that is, all actions related to the administration of accumulated data and its exploitation. In conclusion, it is the environment in which the operation of the database is evaluated.
In general, when talking about database programming, It refers to both the information catalog itself and the data system and its management. For a long time, Oracle Database was the industry leader with an object-relational structure (explained later). However, it now has more and more competitors, highlighting Microsoft SQL Server and MySQL, the latter being one of the most popular open source databases.
Over the last 20 years, databases have become a fundamental piece of any company or institution, from the smallest to the largest. If an organization does not track and analyze all its data – profits, profitability, efficiency of its marketing campaigns, key growth markers, etc. – it does not compete. Databases allow a business to scale, improve its productivity, evaluate its performance, optimize its resources, facilitate the work of its teams, among others.
Database Types
Now that we have answered the question what a database is and seen how it is managed, let’s delve into its structure. Depending on the way data is distributed and stored, databases can be of one type or another.
Likewise, we must consider the evolution of technology, which has also significantly influenced database infrastructure. Let’s see below the most representative models, taking as reference the categorization of Oracle, one of the most popular database management systems on the market.
Relational databases
This data storage and manipulation paradigm was developed in IBM laboratories in the 1970s and quickly became the predominant model in the 1980s. It is based on the distribution of information in tables, composed of columns (fields) and rows (records), and the common characteristics (relationships) that exist between them. SQL is the query, definition, control and data manipulation language designed for the management of this type of databases.
Object database
This model reproduces the way of representing the characteristic information of object-oriented programming, which expanded in the 90s. Therefore, an object-oriented database is usually chosen when working in unison with a programming language that is based on this paradigm such as Java or C#.
Distributed databases
As you can guess from its name, the information in this type of databases andIt is distributed in different locations, physical or digital. In this way, a communication network is necessary to connect the different files with each other.
Data warehouses
The so-called data warehouse in English are a fundamental piece of a business structure and consist of a database system that integrates all the information of an organization in an organized manner for subsequent consultation and analysis. An important aspect of this model is that the data is not deleted and, if a value changes, the new data is added without deleting the previous one. In this way, all operations carried out by an entity are recorded.
NoSQL databases
With the emergence of the Internet and web applications, the need for a more flexible and efficient structure for processing unstructured and semi-structured data became evident. This is where NoSQL databases are born, which, as their name indicates, They abandon SQL as the predominant query language (although they also support it) and the relational data distribution model.
Graph-oriented databases
Just as object-oriented databases represent information as objects, in this case, data is represented as graphs, that is, a structure of vertices and edges that symbolize data sets and the relationships that exist between them.
OLTP databases
Of English Online Transaction ProcessingOLTP refers to the processing of online transactions. Therefore, to the question what is an OLTP database, the answer is one that manages everything that has to do with the structure of users’ transactional data.
It is worth emphasizing that, beyond the types proposed in this post, Databases can be categorized in many ways. We can speak of hierarchical databases, when the structure of their data takes the form of a tree. Likewise, several of these tree structures can be related to each other in a graph-oriented database, among other possibilities.
In any case, the factors that lead a company or institution to opt for one paradigm or another are its specific needs and the technological context in which it finds itself. In recent years, for example, we have seen an explosion of cloud data storage and management, a trend that is still in its genesis. Self-management databases are also in vogue, opening the range of opportunities offered by this technology.
You already know what a database is, how it is structured and what the main types are. Still, do you want more? Have you thought about dedicating yourself to it? With our Big Data, Artificial Intelligence & Machine Learning Full Stack you can specialize, in 9 months and remotely, in everything related to data science: capture and processing, storage, visualization and analysis and much more. Contact us for more information and we will provide you with the study plan that best suits your needs.