What Are Databases?

Databases are organized collections of structured data stored and managed to support efficient retrieval, updating, and administration. The field encompasses the principles, models, and systems used to design, implement, and operate these collections at scales ranging from small embedded applications to distributed data stores serving billions of transactions daily. Databases underpin nearly every modern software system that requires persistent storage of structured information.

The intellectual foundations of databases trace to mathematics and computer science. Edgar F. Codd's 1970 paper published in Communications of the ACM, "A Relational Model of Data for Large Shared Data Banks," introduced the relational model grounded in set theory and first-order predicate logic. That model defined the organizing principles of tabular data, normalization, and structured query operations that dominated database practice for decades, and it earned Codd the ACM Turing Award in 1981.

Relational Databases

Relational database management systems (RDBMS) organize data into tables of rows and columns, enforcing relationships among tables through keys and constraints. The relational model separates the logical structure of data from its physical storage, allowing queries to describe what data is needed without specifying how to retrieve it. Structured Query Language (SQL), standardized by ANSI and ISO, became the universal interface for relational systems. Major commercial and open-source implementations include Oracle Database, IBM Db2, Microsoft SQL Server, PostgreSQL, and MySQL. Relational systems remain the dominant choice for transactional workloads where data consistency and integrity are paramount.

NoSQL and Distributed Databases

The growth of web-scale applications from the early 2000s onward drove the development of non-relational databases, collectively grouped under the label NoSQL. These systems traded strict relational constraints for horizontal scalability and flexible schemas suited to document, key-value, wide-column, and graph data models. The CAP theorem, formalized by Eric Brewer in 2000 and described in his PODC keynote proceedings, frames the trade-offs that distributed databases must navigate among consistency, availability, and partition tolerance. Systems such as Apache Cassandra, MongoDB, and Amazon DynamoDB illustrate the design space: Cassandra prioritizes availability and partition tolerance for wide-column storage, while graph databases such as Neo4j are optimized for querying interconnected data. Many modern architectures combine relational and non-relational stores in a polyglot persistence pattern to match workload requirements.

Query Processing and Transaction Management

Efficient query processing requires the database engine to parse, optimize, and execute queries against potentially large data sets. Query optimizers use statistics about data distribution to select among alternative execution plans, choosing between index scans, full table scans, and join algorithms such as nested loops, hash joins, and merge joins. Transaction management ensures that concurrent operations on shared data satisfy the ACID properties: atomicity, consistency, isolation, and durability. The two-phase commit protocol coordinates distributed transactions across multiple nodes. Logging and recovery mechanisms, including write-ahead logging, allow databases to restore consistent state after failures. IEEE Xplore hosts an extensive body of research on query optimization, concurrency control, and storage engine design that continues to drive advances in database systems.

Applications

Databases have applications in a wide range of fields, including:

• Financial services and banking, where transactional integrity and auditability are critical
• Healthcare records management, supporting electronic health records and clinical data repositories
• E-commerce and retail, managing product catalogs, inventory, and order fulfillment
• Telecommunications, tracking call records, subscriber data, and network topology
• Scientific research, storing genomic sequences, observational data, and simulation outputs
• Geographic information systems, managing spatial data and location-based services