PostgreSQL

In 1986, at the University of California, Berkeley, Professor Michael Stonebraker launched the POSTGRES project. DARPA, the Army Research Office, and the National Science Foundation funded this initiative, which followed INGRES, a relational system developed a few years earlier by the same team. The ambition was to push the boundaries of the relational model in a context where criticism was coming from all sides.

The relational model, introduced by Edgar F. Codd, sparked heated debates in the computer science community. Traditionalists insisted that this type of system could not function properly and that the proposed query languages remained too complex for users. The Berkeley team decided to silence the critics by creating a working prototype.

The first demonstration version appeared in 1987. The technical choices were bold: UNIX as the operating system and the C language for programming. These decisions proved judicious, influenced by exchanges with Ken Thompson, a Berkeley graduate. The team did not hesitate to rewrite entire system components to improve quality, applying an iterative development method that proved successful.

The project’s early years saw intense research and development activity. In 1989, version 1 was distributed to a few external users. Version 2 was released in 1990 with a new rules system. Version 3, published in 1991, brought its share of improvements: support for multiple storage managers, query execution engine optimization, and a redesigned rules system.

POSTGRES found its place in varied applications. It appeared in financial data analysis, aircraft engine performance monitoring, an asteroid tracking database, medical information systems, and geographic applications. Universities adopted it as a teaching tool. In 1992, it became the primary data manager for the Sequoia 2000 scientific project.

The user community doubled in 1993. This growth generated a considerable support burden for the university team. In 1994, two students, Andrew Yu and Jolly Chen, added an SQL interpreter to the system. This evolution gave birth to Postgres95, a more compact and efficient version. The code base was reduced by 25% and performance improved by 30 to 50% on Wisconsin benchmark tests.

In 1996, the project took the name PostgreSQL. This new name emphasized continuity with POSTGRES and the integration of SQL functionalities. This new incarnation marked the beginning of a collaborative and open source approach. The system distinguished itself through its extensibility, which allowed adding new data types and operators without touching the core.

PostgreSQL’s development rests on a minimalist philosophy. The system favors using external tools for functions such as connection management, replication, or backup. This approach promotes the creation of a rich ecosystem of complementary projects and adaptation to technological evolutions like SSD storage or cloud deployment.

The project’s organizational structure reflects its collaborative nature. The core team operates with a minimal charter and no company is capable of employing the majority of its members. This distributed governance ensures that PostgreSQL remains under the control of its community of users and developers.

The emergence of the World Wide Web and mobile applications brought new constraints for traditional relational databases. PostgreSQL adapted by naturally integrating non-relational data types: complex documents, geographic data, GPS information, social network content. This ability to evolve without restructuring constitutes a first-rate technical advantage.

The 2000s saw the growing adoption of PostgreSQL by companies and organizations of all sizes. The system became a serious alternative to proprietary databases from Oracle, IBM, and Microsoft. Its open development model allows for continuous innovation and responsiveness to user needs that traditional vendors struggle to match.

PostgreSQL’s longevity and success result from sound initial architectural choices and stable governance. The system maintains a balance between technical innovation and stability, while preserving its independence from commercial interests. This unique combination explains its current position as the most advanced open source database management system.

Contributions to PostgreSQL come from varied backgrounds: academic researchers, user companies, service providers, independent developers. This diversity nurtures a dynamic of continuous improvement in code, documentation, and associated tools. The project maintains high quality standards while remaining accessible to new contributors.

PostgreSQL continues to evolve to meet modern requirements: large-scale performance, complex data analysis, artificial intelligence. Its extensible design allows the integration of these new functionalities without compromising its consistency or reliability. This adaptability, combined with its technical maturity, makes it a structural technology of contemporary computing capable of competing with historic proprietary solutions.