Database: what is it? Definition and presentation

Computer databases are used in many companies to store, organize and analyze data. Find out everything you need to know about it: what is a database, what is it used for, how does it work, what are the different categories, and what are the best ones.

A database (which we will call BDD for convenience) is a collection of information organized to be easily searchable, manageable and updated. Within a database, data is organized in rows, columns and tables. They are indexed in order to easily find the information sought with the help of computer software. Each time new information is added, the data is updated and eventually deleted.

What is a database? Definition

They are responsible for creating, updating or deleting data themselves. They also search the data they contain at the user’s request.and launch applications from the data.

The databases are used by many companies in all industries. In particular, they are used by airlines to manage reservations. They are used for production management. For medical records in hospitalsor for legal registrations in insurance companies. The broadest databases are generally used by government agenciess, large companies or universities.

How do the databases work?

operating database

Databases are stored as files or sets of files on a magnetic disk, cassette, optical disk or other type of storage device. Traditional (hierarchical) databases are organized by fields, records and files. A field is a single piece of information. A record is a set of fields. A file is a collection of records.

For example, a telephone directory is the equivalent of a file. It contains a set of records, and each record has three fields: name, address and phone number. Other examples include product catalogues or inventories.

The ability to consult or modify a database (read or write) is granted to the various users by a database manager. Databases are mainly present in the largest mainframe systems, but they are also found in smaller distributed workstations and other midrange systems such as IBM AS/400s or even personal computers.

History of the databases

History of the databases dates back to the 1960swith the advent of network databases and hierarchical databases. In the 1980s, object-oriented databases appeared. Today, the predominant databases are SQL, NoSQL and cloud databases.

It is also possible to classify databases according to their content: bibliographic, text, numbers or images. However, in computer science, databases are generally classified according to their organizational approach. There are many different types of databases: relational, distributed, cloud, NoSQL… Here are the different types of databases.

What are the different types of databases

sample database

In the case of a large database, multiple users must be able to manipulate the information it contains quickly and at any time. In addition, large companies tend to accumulate many independent files with linked files or even overlapping data. When analyzing data, it is necessary for data from multiple files to be able to be linked. This is why different types of databases have been developed to meet these requirements : text-oriented, hierarchical, network, relational, object-oriented

Hierarchical database

Hierarchical databases include oldest databases. Within this category, records are organized in a tree structure. Each level of records flows into a set of smaller categories.

Network database

Network databases are also among the oldest. Rather than providing unique links between different datasets at various levels, network databases create multiple links between the sets by placing links, or pointers, to one set of records or another. The speed and versatility of network databases has led to a massive adoption of this type of database within companies or in the e-commerce field.

Text-oriented database

A text-oriented database, or flat file database, takes the form of a file (a table) in .txt or .ini format. A flat file is a text file, or a file combining text with a binary file. Most PC databases are text-oriented databases.

SQL database (relational)

Relational databases were invented in 1970 by E.F. Codd of IBM. They are tabular documents in which the data is defined to be accessible and can be reorganized in different ways.

Relational databases consist of a set of tables. Within these tables data are categorized. Each table has at least one column corresponding to a category. Each column contains a certain amount of data corresponding to that category.

The standard API for relational databases is Structured Query Language (SQL). Relational databases are easily extensible, and new categories of data can be added after the original database is created without the need to modify all existing applications.

Distributed database

A distributed DB is a database of which some portions are stored in multiple locations physical. Processing is distributed or replicated between different points in a network.

Distributed databases may be homogeneous or heterogeneous. In the case of a homogeneous distributed database system, all physical locations run on the same hardware and run the same operating system and database applications. In contrast, in the case of a heterogeneous distributed database, the hardware, operating systems and database applications may vary between the different physical locations.

Cloud database

In this context, it is optimized or directly created for virtualized environments. It can be a private cloud, a public cloud or a hybrid cloud.

The Cloud databases offer several advantages such as the ability to pay for storage capacity and bandwidth based on usage. In addition, it is possible to change the scale on request. These databases also offer higher availability.

NoSQL database

NoSQL databases are useful for large distributed datasets. Indeed, relational databases are not designed for Big Data, and too large datasets can cause performance problems.

If a company needs to analyze large amounts of unstructured dataFor data stored on multiple virtual cloud servers, the NoSQL database is ideal. With the rise of Big Data, NoSQL databases are more and more used.

Object-oriented database

The objects created using object-oriented programming languages are usually stored on relational databases. However, in reality, object-oriented databases are more suitable for storing this type of content.

Rather than being organised around actions, object-oriented databases are organized around objects. Similarly, instead of being organized around logic, they are organized around data. For example, a multimedia record within a relational DB can be defined as a data object rather than an alphanumeric value.

Graphically oriented database

A graph-oriented database, or graph, is a NoSQL database type using graph theory to store, map and query data relationships. Plot databases consist of nodes and edges.

Each node represents an entity, and each edge represents a connection between the nodes. Graph databases are gaining popularity in the field of interconnection analysis. For example, companies can use a graphical database to Undermine customer data from social networks.

Increasingly, formerly separate databases are being combined electronically into larger collections. called Data Warehouses.. Companies and governments then use data mining software to analyze different aspects of the data. For example, a government agency may do this to investigate a company or individual who has purchased a large amount of equipment, even if the purchases are spread across the country or spread among several subsidiaries.

Can relational and non-relational databases coexist?

sql nosql database

It may be tempting to see NoSQL databases as a replacement for relational databases. However, in reality, there may be room for both types of technologies in most companies.

SQL and NoSQL databases handle information in a different way and support different types of workloads. Rather than taking the place of relational databases, NoSQL databases allow companies to aim for new goals and meet new challenges.

It’s important to know in which situation to use a NoSQL databaseand in which situation it is preferable to use a relational DBD. Depending on the nature of the workload and the underlying data, a relational or NoSQL solution should be used.

For example, the databases NoSQL is more suitable for large sets data that are frequently exposed to new information, in which the records have variable structures that do not fit well with relational models.

Another reason why non-relational databases will not replace relational databases is that relational database vendors have enhanced their products to fit Big Data applications. Not only will these two technologies still have their place in businessbut there is also room for both of them to continue to grow on their own.

How the cloud transforms databases

cloud database

Cloud databases differ from traditional databases by the fact that they are reside in the public, private or hybrid cloud and not on internal servers. There are two distinct cloud database environments: traditional databases and databases as a service (DBaaS).

A traditional cloud database can be launched on a company’s IT infrastructure. through a virtual machine. The maintenance and management of the database is carried out by the company’s IT staff.

In the case of a DBaaS, the company subscribes and relies on the physical infrastructure of the cloud service provider. The provider is responsible for maintaining the physical infrastructure and the base, while the customer is responsible for managing the content and operations of the DB. Some vendors also offer to take over management, a very useful option for small businesses that need a database but do not have the IT expertise.

The contributions of cloud databases

Cloud databases bring many advantages compared to a traditional database. First of all, they eliminate the physical infrastructure. The physical infrastructure is provided remotely by the cloud provider who also takes care of maintenance and ensures availability.

The second advantage is economical since cloud databases reduce costs. The elimination of physical infrastructure reduces expenses, and implies a reduction in the number of staff required and the electricity bill. The cloud also saves physical space.

A DBaaS has even more advantages. The user enjoys a instant scalabilityThis allows it to increase or reduce the capacity of its database according to its needs. The service provider quickly takes care of increasing the capacity allocated to the customer. Installing one on site would take several months to achieve the same result, with the need to install new servers and other communication resources.

The importance of ALS

In addition, the service level agreements (SLAs) require the supplier to guarantee DB performance, availability and response time. If this level of service is not met, the customer is reimbursed. The user of a DBaaS also benefits from the service provider’s expertise, something that would be very difficult to find from employees due to the high costs of such a qualification. The DBaaS provider has thousands of customers, so it is not a problem for it to recruit the elite of database administrators.

DBaaS providers also ensure that their database software, operating systems and all technologies related to their infrastructure are state-of-the-art. This allows them to remain competitive. Functionality and security are regularly updated. For the customer, this is the assurance of take advantage of the latest technologies.

The different types of databases

dbaas database

To avoid interruption of operations in the event of a failure, DBaaS providers use mirror server images and have several data storage complexes at their disposal. In the event of a failure, operations are transferred to a backup data center and the customer is not aware of anything. A final advantage is that, thanks to technological advances and fierce competition in the market, the price of cloud services is gradually decreasing. DBaaS are therefore increasingly accessible.

Like traditional databases, the Cloud databases fall into two categories : relational and non-relational. The differences between these two categories and their respective advantages, mentioned above, are valid for cloud databases.

It is quite possible to migrating on-premise databases to the cloud. This approach has several benefits. It eliminates the need for physical servers and storage infrastructure, and saves the company from having to find database experts. Migration can also improve the efficiency of data processing, especially when analytical tools and other applications are also in the cloud.

Cloud Migration

Of course, migrating a database to the cloud also allows you to save money by reducing costs. The IT team is reduced, so fewer employees need to be paid. On the other hand, the prices in the cloud are getting lower and lower, which makes it possible to take advantage of ever more tempting offers. In addition, cloud service providers offer their customers the option of paying only for the resources they consume. This is the pay-as-you-go model.

A migration to the cloud may allow for more adoption of SaaS (software as a service) solutions in the future. These cloud services simplify access to information over the Internet. Storage consolidation can be another advantage of a migration to the cloud. For example, the databases of the different departments of a large company can be combined in the cloud into a single database management system.

In terms of structural operation and design, a cloud database is similar in every way to an on-premise structure. The main difference is where it resides. An on-premise database is connected to the various local users of the company. via a LAN type local area network. A cloud DBD resides on the servers of a cloud provider or DBaaS. It is only accessible via the internet.

In both cases, the database is accessed through direct queries or API calls, and the database responds in the same way. However, response time may vary. An on-premises infrastructure is logically a little faster than one installed in the cloud, because every interaction on a cloud base has to be transferred to the internet. However, the difference is minimal.

DBMS (database management system) or DBMS and RDBMS: software that provides access to databases

dbms rdbms database

To access the databases, a DBMS (database management system) is used, i.e. a DBMS or RDBMS type software. A DBMS is a database management system. It is used to define, manipulate, retrieve and manage the data stored in the DB. The DBMS extracts information from the DB on request of the user, according to the requests made. To perform a query, the user can enter a keyword, or perform a sort command.

The power of a DBMS is its ability to defining new relationships from basic relationships given by the tables to answer queries. Typically, the user enters a series of characters, and the computer searches for the corresponding sequences to provide the user with the source materials in which these characters appear. For example, a user can search for all records containing a field related to a person with the last name Smith.

History of the DBMS

The term Database is increasingly being used as an abbreviation for for Database Management System. There are many different DBMS. Some are small systems that can be run on a personal computer, others are huge systems requiring a mainframe.

The DBMS were invented in the 1960s. to support hierarchical databases. Early systems were organized sequentially (alphabetically, numerically, or chronologically). It was not until the advent of direct-access storage devices that data could be accessed randomly through indexes. Among the best-known DBMSs are the IBM Information Management System and the CA Integrated Database Management System.

An RDBMS is a relational database management system. This type of software has been developed in the 1970s based on the relational model. Even today, it remains the most popular way to manage a comic book. The best known RDBMSs are Microsoft SQL Server, Oracle Database, IBM DB2 and MySQL.

What is SQL or Structured Query Language?

database sql

SQL (Structured Query Language) is a standard programming language used to manage relational databases and perform various operations on the data they contain. This language was created in the 1970s, and continues to be used regularly by database administrators. Developers who write data integration scripts and data analysts also use it to run analytical queries.

An official SQL standard was adopted by ANSI (American National Standards Institute) in 1986. In 1987, the ISO (International Organization for Standardization) adopted it in turn. This standard has been updated more than six times since then. The most recent version is SQL:2011.

The use of SQL allows toe modifying the structures of the tables and indexes of the databases. The user can add, update and delete rows of data and retrieve subsets of information. This information can then be used for analytical applications or transaction processing. A rSQL header takes the form of a written command. The most frequently used commands are select, add, insert, update, delete, create, alter and truncate.

SQL commands and the different tools

The SQL commands are divided into several different types. The DML (Data Manipulation Language) vocabulary is used to retrieve and manipulate data. DDL (Data Definition Language) is used to define and modify database structures. Transaction control is used to manage transactions to ensure that they are completed or cancelled in the event of problems or errors. Finally, security statements are used to control access to databases and create a system of permissions for different users.

Companies use RDBS or database management systems developed around proprietary or open source SQL. Some of the best known include Microsoft SQL Server, Oracle Database, IBM DB2, SAP HANA, SAP Adaptive Server, Oracle MySQL and PostgreSQL. Despite the standardization of SQL, most vendors are using proprietary extensions for procedural programming and other functions. For example, Microsoft offers a Transact-SQL extension set (T-SQL) and Oracle has its extended version (PL/SQL). In fact, the variants of the various vendors are not compatible with each other.

Recently, SQL-on-Hadoop query engines have been used for Big Data architectures developed around Hadoop systems. These engines make it possible to exploit Big Data without having to use a more complex language. One example is the MapReduce programming environment for the development of batch processing applications. In total, there are more than a dozen SQL-on-Hadoop tools offered by the various Hadoop distribution vendors and other vendors. Note that the Apache Spark processing engine integrates its own SQL compatible Spark module.

What is a DataBase Administrator?

dba database administrator

A Database Administrator or DataBase Administrator is the person responsible for maintaining such an environment. The design, implementation, maintenance of the system and the implementation of rules. It must also train the company’s employees in the management and use of the DB.

As a general rule, a DBA has a background in computer sciences and professional experience with a specific or various databases. He must also have experience with major database management products such as SQL, SAP or Oracle.

What are the best databases at the moment?

open source database

The choice of a good database is very important for your business. It must be easy to use, secure, with a good development follow-up can increase productivity. It is therefore essential to study the advantages and disadvantages of each of them. The database you choose must fit your company’s ecosystem.

Several key questions need to be asked How many items will you have to manage, what is the reaction time expected by your customers, how many customers do you have, how will it adapt if your number of customers and transactions increases, how will you monitor it to avoid down times, do you need a relational or NoSQL database, and how will it behave in case of crash or problem?

Currently, the market is dominated by DB2, SQL Server, Oracle and IBM.. On Windows, SQL Server is usually the preferred DB, while Oracle and DB2 are the most popular on Mainframe/Unix or Linux ecosystems. To help you find the one that’s right for you, here is our selection of the best databases.

Microsoft SQL Server, the Windows database

Developed by Microsoft, SQL Server, it is Windows-only compatible. This database is simple to master, and many people are trained in it. The integration with Microsoft Azure has increased its flexibility and performance. In addition, the cloud now allows the administration of information from other servers, which makes it more useful.

Oracle, the most popular database for Linux/Unix

The Oracle database can run on almost any system. It is very popular, and from many people are trained to master it. In addition, it offers numerous tools for monitoring and administration.

IBM DB2, the most popular mainframe database

After Oracle, IBM DB2 is the second most used on Unix/Linux ecosystems. For Mainframe, it’s the most popular choice. Again, many people are trained to use it, even though it has fewer followers than Oracle.

Teradata, the best database for Big Data

Teradata is specifically designed for Big Data. Its data storage and analysis capacities are colossal. For a Big Data strategy, this is undoubtedly the best option available to you.

SAP Sybase, the former market leader

In the past, this database was very popular and largely dominated the market. Today, it is no longer used as much, but remains a very relevant solution in terms of scalability and performance.

Informix, a database acquired by IBM

Just like SAP Sybase, Informix has lost some of its superb in the late ’90s. Following a series of bad management decisions, it was eventually bought out by IBM.

It no longer exists today in its original form. Even so, its foundations are still used by some IBM tools and applications.

MySQL: the best Open Source relational database

mysql

In the very particular category of Open Source relational databases, there are three main references: MySQL, MariaDB, and PostgreSQL. These three management systems have several things in common: a very active support community, an open code allowing users to modify them as they wish according to their needs, and free of charge (for developers).

MySQL is a relational database management system (RDBMS) based on SQL (Structured Query Language). This RDBMS is compatible with all : Linux, UNIX, and Windows. It can be used with many applications, but it is most often associated with web applications.

MySQL was originally created by the Swedish company MySQL AB. However, in 2008, Sun Microsystems acquired it before being bought out by Oracle itself in 2010. Developers can still use MySQL under the GPL, but companies must purchase a commercial license from Oracle.

MySQL is also a component of the LAMP enterprise software suitea web development platform using Linux as the operating system, Apache as the web server, MySQL as the RDBMS and PHP as the object-oriented scripting language.

PHP is indeed the most popular language for web development. It is a free, open source and server-side language, which means that the code is executed on the server. The combination between PHP and MySQL allows you to create any type of site, from a simple contact form to a corporate portal.

PHP also allows to connect a web script to a MySQL database. This is an essential step, indispensable for being able to make queries.

Free database software

There are various free management software packages available. Some are simply free duplicates of the famous Microsoft Excel (Open Office Base), others are more complex to use and take the role of DBMS. This is the case of SQLite, a database engine, MariaDB, MySQL, MaxDB, Ingres or PostgreSQL.

What are the best NoSQL databases?

Among NoSQL databases, there are many sub-categories. Each of these categories is distinguished by specific characteristics. Here are the four main groups of NoSQL databasesand the best in each of these categories.

NoSQL key-value oriented databases

These databases are ideal for access the data via a key. The specificity is that the data can be stored without defining a specific schema. These databases are very efficient for reading and writing, and are designed to be massively scalable and offer an extremely fast response time.

Elements are usually stored within complex structures such as the BLOB. References in this category of databases are Redis, Riak, Oracle NoSQL and Microsoft Azure Table Storage. Radis is freebase and open source, Riak is entirely dedicated to value keys and ideal for document storage and search functionalities.

Document-oriented NoSQL databases

This category of databases is based on various formats (JSON, XML) and offers the ability to change the schema without having to shut down the database. Developers can download indexed documents and access them through the database storage engine. The flexibility of these databases makes them very versatile.

Some of the best document-oriented NoSQL databases include Mongo DB and Couchbase Server. Mongo DB is one of the most popular databases at the moment, all categories included. It supports both structured and unstructured data and has excellent performance and scalability. Many people are trained to master it.

Couchbase Server is an Open Source database licensed under Apache. Its main advantage is its very intuitive management console allowing easy access to large amounts of data. However, it does not guarantee 100% data completeness.

One of the best document-oriented NoSQL databases, we can also quote Mark Logic Server. Its data integrity and XML, JSON and RDF compatibility make it a reference. Mark Logic Server is compatible with Windows, Solaris, Red Hat, Suse, CentOS, Amazon Linux and Mac OS. Finally, we can mention Elastic Sarche, RavendDB, Apache Jena and Pivotal GemFire.

Column oriented NoSQL databases

NoSQL databases oriented to column represent the value of the data as a columnwhich allows the user to map keys and values and group them into structures. These databases are mainly used in environments where it is necessary to access many columns with many rows. They are especially useful for event processing and analysis, content management and data analysis.

Some of the best column-oriented NoSQL databases include Apache Cassandra, an engine created by Facebook and now distributed for free. Cassandra is recommended for databases with huge amounts of data.

It should be noted that a enterprise version called Datastax Enterprise is also available. Cassandra supports ASCII, bigint, BLOB, Boolean, counter, decimal, double, float, int, text, timestamp, UUID, VARCHAR and vary. The other reference in this category is Apache Hbase, designed to support multiple real-time read and write access to huge amounts of data.

Graphically oriented NoSQL databases

Graphically oriented NoSQL databases are focused on properties and relationships that unite them. They use graph theory to connect databases. Each element is connected to its adjacent element. These databases are recommended if your data is interrelated, such as on social networks, in fraud detection or real-time updates.

References in this category are Neo4J and Infinite graph. Neo4j supports data integration, high availability, and clustered scaling. Its administration panel is also very good. Infinite Graph is a licence-only database compatible with MacOS, Linux and Windows.

Be the first to comment

Leave a Reply

Your email address will not be published.


*