データベース

Database — Grokipedia Fact-checked by Grok 2 months ago Database Ara Eve Leo Sal 1x A database is an organized collection of structured information, or data, typically stored electronically in a computer system and controlled by a database management system (DBMS). [1] [2] The DBMS acts as an interface between the database and users or applications, enabling efficient storage, retrieval, updating, management, and secure access to data while enforcing rules for data integrity, consistency, and security. [1] [3] Databases support the handling of large volumes of information across diverse applications, from simple record-keeping to complex analytics, by providing mechanisms for data organization, querying (often via Structured Query Language, or SQL), and multi-user access. [1] They ensure data remains accurate and reliable through built-in constraints and reduce redundancy by maintaining a centralized, trustworthy repository. [3] [2] Databases are categorized into several main types based on data structure and use case. Relational databases organize data into tables with rows and columns, linked by keys, and rely on SQL for querying; they excel with structured data and remain dominant for applications requiring precise relationships and transactions. [1] [3] NoSQL (non-relational) databases accommodate unstructured or semi-structured data without fixed schemas, offering flexible models such as document stores (e.g., JSON-based), key-value stores, wide-column stores, graph databases (focused on relationships between entities), and in-memory stores for high-speed access. [1] [4] Other variants include object-oriented, distributed, cloud-based, and emerging autonomous databases that automate management tasks using machine learning. [1] These systems are foundational to modern computing, powering websites, mobile applications, enterprise operations, e-commerce, fraud detection, recommendation engines, gaming, and AI-driven analytics by enabling scalable storage, real-time processing, and data-driven decision-making. [2] [3] [4] Databases have evolved from early navigational models in the 1960s–1970s to relational dominance in the 1980s, the rise of NoSQL in the 21st century for big data and web-scale needs, and recent cloud and self-managing innovations for greater efficiency and agility. [1] [2] [3] Overview Definition A database is an organized collection of structured data, typically stored electronically in a computer system. [1] This structure enables efficient storage, retrieval, and manipulation of information, often modeled in rows and columns within tables for relational databases, though other formats exist for non-relational types. [1] A database is usually controlled by a database management system (DBMS), which acts as an interface between the data and users or applications, facilitating administrative tasks such as performance monitoring, backup, and recovery. [1] The primary purpose of a database is to manage, store, retrieve, and update large volumes of information reliably and efficiently. [2] Databases support multi-user access, allowing simultaneous queries and modifications while maintaining data integrity (accuracy and completeness), security (through controls like role-based access), and consistency (ensuring data remains synchronized and reliable across operations). [3] These features make databases essential for handling complex data needs beyond what simpler tools can provide. Unlike traditional file systems, which store data in independent files without built-in mechanisms for enforcing relationships, concurrency control, or integrity constraints, databases offer centralized management to reduce redundancy and errors. [5] Similarly, while spreadsheets are suitable for small-scale, single-user tasks involving calculations and basic data manipulation, databases are designed for larger, structured datasets with advanced querying and multi-user support. [6] This distinction enables databases to serve as robust foundations for modern applications requiring scalability and reliability. Key characteristics Databases are distinguished by several essential characteristics that enable them to manage large volumes of structured data effectively, far beyond simple file storage systems. A primary trait is their self-describing nature , where the system stores not only the data but also metadata describing the structure, relationships, and constraints, allowing the database to operate independently of specific applications. [7] [8] Data independence is another key property, separating the logical organization of data from physical storage and application programs; changes to storage structures or access methods do not require modifications to programs that use the data. [7] [9] Databases support persistence through durability, ensuring that committed changes survive system failures, and multi-user access , enabling concurrent operations while maintaining consistency via concurrency control mechanisms. [10] [8] They incorporate controlled redundancy , ideally storing each data item in one place to minimize duplication while allowing necessary replication for performance, and provide efficient querying through structured organization and optimized retrieval methods. [9] Core design goals include data integrity , enforced through constraints and ACID properties (Atomicity, Consistency, Isolation, Durability), which guarantee that transactions complete reliably and the database transitions only between valid states. [10] Security restricts unauthorized access via user privileges and authentication, while consistency ensures data remains accurate and coherent across operations. [11] [9] Databases encompass various types, such as relational and NoSQL, which adapt these characteristics to different use cases. Importance and applications Databases are the backbone of modern digital systems, storing, organizing, and managing data in ways that make it accessible, secure, and actionable across virtually every computing application. [12] They power websites, mobile apps, enterprise platforms, and real-time analytics, enabling concurrent access, data consistency, and integration with applications at scale. [12] Without databases, modern software systems could not efficiently handle the vast amounts of data generated daily, nor support the performance and reliability demanded by contemporary applications. [13] Databases enable data-driven decision-making by facilitating analysis that identifies trends, patterns, and predictions, which helps organizations operate with greater confidence and efficiency. [14] In business, they underpin critical functions such as customer relationship management, inventory tracking, financial reporting, fraud detection, and transaction processing in banking and e-commerce. [12] [14] For example, relational databases manage structured customer and order data across retail locations, while other types support caching for faster website performance or storing user-generated content on social platforms. [12] Databases also manage large and complex datasets in scientific research and academia—including simulation models of real-world entities and experimental results—supporting knowledge advancement and collaborative analysis. [15] [16] In government and public sectors, they support administrative operations, public service delivery, and evidence-based policy through secure, scalable information management. [17] Overall, databases enable scalability, data integrity, and security, making them indispensable for information management across various sectors including business, research, and public administration. [12] [14] [17] History 1960s: Navigational databases In the 1960s, the first computerized database management systems emerged, adopting navigational approaches to data access. These systems required programmers to traverse data structures explicitly by following pointers or links between records, typically processing one record at a time in a "record-at-a-time" manner. This navigational paradigm contrasted with later declarative methods and was implemented in both hierarchical and network models. [18] [19] A pioneering effort was Charles W. Bachman's Integrated Data Store (IDS), developed at General Electric. Functional specifications for IDS were completed in early 1962, with a prototype operational by the end of that year and a higher-performance version finished in 1964. IDS introduced the network data model, representing relationships between records as a graph and enabling programmers to navigate these links using commands to retrieve and process individual records. Bachman described the programmer's role as a "navigator" through these interconnected structures, and IDS supported features such as data independence, metadata management, and transaction processing enhancements by 1965. [19] [20] Bachman's work heavily influenced the Conference on Data Systems Languages (CODASYL) Database Task Group , which formed to standardize database approaches. The group's 1969 report defined a network model standard, building directly on IDS concepts and including data definition and manipulation languages, schemas, and security features such as privacy locks. The CODASYL network model allowed flexible many-to-many relationships through sets, with records accessed via primary keys, set navigation, or sequential scanning. [19] [20] Concurrently, IBM developed the Information Management System (IMS), a hierarchical database. Development began in 1966 in collaboration with Rockwell for NASA's Apollo program needs, with the initial version shipped in 1967 and delivered to NASA in 1968. IMS organized data in tree-like structures with parent-child relationships, requiring navigation down the hierarchy to access dependent records. It was designed for high-volume transaction processing and efficient management of complex inventories, such as rocket parts tracking, and was commercially announced in 1968