Databases are used for storing, maintaining and accessing any sort of data. They collect information on people, places or things. That information is gathered in one place so that it can be observed and analyzed. Databases can be thought of as an organized collection of information.
A database is information that is set up for easy access, management and updating. Computer databases typically store aggregations of data records or files that contain information, such as sales transactions, customer data, financials and product information.
Businesses use data stored in databases to make informed business decisions. Some of the ways organizations use databases include the following:
The relational database, combined with the growth of the internet beginning in the mid-1990s, led to a proliferation of databases. Many business and consumer applications rely on databases.
E.F. Codd created the relational database while at IBM. It became the standard for database systems because of its logical schema, or the way it is organized. The use of a logical schema separates the relational database from physical storage.
Relational databases were developed in the 1970s. Object-oriented databases came next in the 1980s. Today, we use Structured Query Language ( SQL ), NoSQL and cloud databases .
Databases were first created in the 1960s. These early databases were network models where each record is related to many primary and secondary records. Hierarchical databases were also among the early models. They have tree schemas with a root directory of records linked to several subdirectories.
Types of databases
There are many types of databases. They may be classified according to content type: bibliographic, full text, numeric and images. In computing, databases are often classified based on the organizational approach they use.
Some of the main organizational databases include the following:
Relational. This tabular approach defines data so it can be reorganized and accessed in many ways. Relational databases are comprised of tables. Data is placed into predefined categories in those tables. Each table has columns with at least one data category, and rows that have a certain data instance for the categories which are defined in the columns. Information in a relational database about a specific customer is organized into rows, columns and tables. These are indexed to make it easier to search using SQL or NoSQL queries.
Relational databases use SQL in their user and application program interfaces. A new data category can easily be added to a relational database without having to change the existing applications. A relational database management system (RDBMS) is used to store, manage, query and retrieve data in a relational database.
Typically, the RDBMS gives users the ability to control read/write access, specify report generation and analyze use. Some databases offer atomicity, consistency, isolation and durability, or ACID, compliance to guarantee that data is consistent and that transactions are complete.
Distributed. This database stores records or files in several physical locations. Data processing is also spread out and replicated across different parts of the network.
Distributed databases can be homogeneous, where all physical locations have the same underlying hardware and run the same operating systems and database applications. They can also be heterogeneous. In those cases, the hardware, OS and database applications can be different in the various locations.
Cloud. These databases are built in a public, private or hybrid cloud for a virtualized environment. Users are charged based on how much storage and bandwidth they use. They also get scalability on demand and high availability. These databases can work with applications deployed as software as a service.
NoSQL. NoSQL databases are good when dealing with large collections of distributed data. They can address big data performance issues better than relational databases. They also do well analyzing large unstructured data sets and data on virtual servers in the cloud. These databases can also be called non-relational databases.
Object-oriented. These databases hold data created using object-oriented programming languages . They focus on organizing objects rather than actions and data rather than logic. For instance, an image data record would be a data object, rather than an alphanumeric value.
Graph. These databases are a type of NoSQL database. They store, map and query relationships using concepts from graph theory. Graph databases are made up of nodes and edges. Nodes are entities and connect the nodes.
These databases are often used to analyze interconnections. Graph databases are often used to analyze data about customers as they interact with a business on webpages and in social media.
Graph databases use SPARQL, a declarative programming language and protocol, for analytics. SPARQL can perform all the analytics that SQL can perform, and can also be used for semantic analysis, or the examination of relationships. This makes it useful for performing analytics on data sets that have both structured and unstructured data. SPARQL lets users perform analytics on information stored in a relational database, as well as friend-of-a-friend relationships, PageRank and shortest path.
