Timeline of Relational Databases
Relational databases from Edgar F. Codd’s 1970 paper on the relational model to DuckDB 1.0 in 2024: query languages, optimization, transactions, open-source systems, and distributed and analytical designs.
- Category
- Engineering & Technology
- Tag
- computing
- Events
- 20 events 1970–2024
- Generated
- by gpt-6-astra
1970: Codd proposes the relational model
Edgar F. Codd published "A Relational Model of Data for Large Shared Data Banks" while working at IBM, expanding on a 1969 IBM research report. He proposed describing data through relations rather than navigation paths, separating logical queries from the details of physical storage.
Wikipedia1974: Chamberlin and Boyce introduce SEQUEL
Donald Chamberlin and Raymond Boyce presented SEQUEL (Structured English Query Language), developed for IBM’s relational research. The language let users state the data they wanted without specifying a record-by-record retrieval procedure; it later became Structured Query Language (SQL).
Wikipedia1974: Berkeley builds the Ingres prototype
Michael Stonebraker and Eugene Wong’s team at the University of California, Berkeley developed an early Ingres prototype. The university project tested relational ideas independently of IBM’s System R and influenced later database systems and companies.
Wikipedia1976: Chen introduces entity–relationship modeling
Peter Chen published a model for describing entities, their attributes, and relationships. The approach gave database designers a conceptual representation of an application’s data before choosing tables and implementation details.
Wikipedia1979: System R researchers publish cost-based optimization
Patricia Selinger and colleagues described how System R selected query execution plans using estimates of their costs. The optimizer considered choices such as access paths and join order, showing how declarative queries could be translated into efficient execution.
Wikipedia1979: Oracle releases a commercial SQL database
Relational Software, later renamed Oracle, shipped Oracle Version 2 as a commercial implementation of Structured Query Language. The release helped bring relational querying from research projects into business software.
Wikipedia1981: IBM commercializes relational querying with SQL/DS
IBM released Structured Query Language/Data System, bringing technology developed through System R into a commercial database product. Customers could use relational queries on supported IBM mainframe systems.
Wikipedia1983: IBM announces Db2 for mainframes
IBM announced Db2 for its mainframe operating-system line, extending its commitment to relational databases. The product helped establish Structured Query Language as a common interface for enterprise data management.
Wikipedia1983: Härder and Reuter name transaction guarantees
Theo Härder and Andreas Reuter grouped atomicity, consistency, isolation, and durability under the acronym ACID. The vocabulary clarified what reliable database transactions should provide, although products and isolation settings could offer different practical guarantees.
Wikipedia1986: SQL receives an American national standard
The American National Standards Institute adopted a standard for Structured Query Language, followed by an international standard the next year. Standardization provided a shared reference even as database vendors continued to implement their own extensions.
Wikipedia1986: POSTGRES explores extensible database design
Michael Stonebraker’s Berkeley team began implementing POSTGRES as a successor to Ingres. The project investigated richer data types and rules, seeking to support complex applications beyond the fixed types offered by many existing relational systems.
Wikipedia1992: SQL-92 expands the common language
The SQL-92 standard substantially enlarged Structured Query Language, including more explicit join syntax and richer constraint definitions. Its conformance levels gave implementers a staged target rather than requiring every database to support the whole language immediately.
Wikipedia1995: Widenius and Axmark create MySQL
Michael Widenius and David Axmark made the first internal release of MySQL, a database server with a Structured Query Language interface; public releases followed in 1996. Its availability and integration with web-development tools later made it a common component of open-source website stacks.
Wikipedia1997: PostgreSQL 6.0 ships under a new name
PostgreSQL 6.0 followed Postgres95, which had added a Structured Query Language interpreter to Berkeley’s POSTGRES code. The new name connected its research ancestry to the community-maintained database project.
Wikipedia2000: SQLite embeds a database in applications
D. Richard Hipp released SQLite as a database engine linked directly into an application. Eliminating a separate database-server process made relational storage useful in embedded software and local application files as well as server deployments.
Wikipedia2005: C-Store presents a column-oriented design
The C-Store research team described a relational engine that stored values by column rather than by row. Compression and execution tailored to analytical queries showed how a different physical layout could improve workloads that scanned a few columns across many records.
Wikipedia2012: Spanner joins global distribution with transactions
Google published Spanner’s design for a globally distributed database with externally consistent transactions. Its TrueTime interface exposed clock uncertainty, allowing transaction coordination to account for time bounds instead of assuming perfectly synchronized clocks.
Wikipedia2014: Amazon previews Aurora
Amazon Web Services announced Aurora as a MySQL-compatible managed database in preview. The design separated database processing from a distributed storage service, shifting replication and durability work into the storage layer; general availability followed the next year.
Wikipedia2017: CockroachDB releases version 1.0
Cockroach Labs released a distributed database with a Structured Query Language interface and automatic replication. CockroachDB combined transactions with distribution across machines, aiming to keep applications running through machine and infrastructure failures.
Wikipedia2024: DuckDB reaches version 1.0
DuckDB reached its first stable release as an analytical database running inside an application process. Developed originally at Centrum Wiskunde & Informatica in the Netherlands, it combined column-oriented execution with direct querying of local data files.
Wikipedia