- +41 41 201 88 44
- [email protected]
- Mon-Fri 8am - 6pm
A Brief History on Data Architecture and How it has evolved
By Swiss Himmel
A database system’s data architecture establishes the types of data that will be gathered as well as how they will be utilised, processed, and stored. For instance, the Data Architecture provides guidance on the integration process for data integration. Modern computers would be substantially slower and clumsier if the transition from the programming paradigm to the Data Architecture paradigm hadn’t occurred.
Data integration was not even a notion when computers were first invented; instead, simple programmes were written to solve certain computer issues. Each programme was kept separate from the others. Program processing was the main focus from the 1940s until the early 1970s. The idea of an architectural structure for data was typically not given much thought, if any. The primary purpose of a programmer was to direct a computer to carry out particular tasks that aided an organization’s immediate objectives. Computers were not utilised for long-term data storage; instead, they were only used to access data that was “required for the programme.” Writing systems that could retrieve certain information was necessary for data recovery, and this took time and money.
A Paradigm Shift from Programming to Database Architecture
A relational method for organising data was described in a study by Edgar F. Codd in 1970 titled “A Relational Model of Data for Large Shared Data Banks.” The mathematics of set theory served as the foundation for Codd’s theory, which also included a set of guidelines to ensure that data was kept with the least amount of duplication possible. His strategy was successful in producing database architectures that improved computers’ efficiency. Prior to Codd’s work, COBOL programmes and the majority of others used hierarchical data structures.
In 1976, Peter Chen introduced “entity/relationship modelling,” now more frequently referred to as “data modelling,” in a paper he wrote when he was a student at MIT entitled The Entity-Relationship Model-Toward a Unified View of Data. His method used visual representations of data structures. Oracle unveiled the first relational database management system (RDBMS) intended for commercial use two years later.
Computer experts started to understand that these data structures were more dependable than programme structures. The system’s midsection was redesigned, and the processes were separated from one another to maintain this stability (similar to the way programmers kept their programmes isolated). Data buffers were crucial to this redesign’s success.
The Paradigm Shift's Loss of Assumptions
Three fundamental assumptions had to be eliminated in order for Data Architecture to advance:
First Assumption
All programmes ought to be kept separate from one another. Duplications of programme codes, data definitions, and data inputs resulted from this isolationist mentality. Unneeded duplication was addressed using Codd’s relational method. His paradigm distinguished between the physical information storage and the database’s schema, or layout (becoming the standard for database systems).
Second Assumption
Since input and output are equal, matched pairs should be used in design. Currently, data processing speeds on both input and output devices might vary greatly. The notion that both would work at the same pace is significantly different from this. The understanding that output may and should be treated differently from input was brought on by the usage of buffers. The distinctions between data makers and consumers were made clear by Peter Chen’s inventions.
Third Assumption
A company’s computer programmes ought to match the way it is set up. The idea that “programmes” should replicate a company’s organisational structure eventually changed with the usage of buffers and a relational database. The function of giving organisations a helpful framework to adhere to while gathering and processing information was taken over by the more flexible databases.
Since its inception, data architecture has undergone a complete transformation. It is likely that this transformation will continue well into the future as a result of newer trends like the Internet of Things, cloud computing, microservices, advanced analytics, machine learning, and artificial intelligence, as well as emerging technologies like blockchain.
Address
Steinentorstrasse 35 , 4051 Basel, Switzerland
Copyright 2022 by SwissHimmel All Right Reserved.