Skip to content
Get App

History of Databases

Explores the history of databases, from Sumerian tablets and flat files to IBM's hierarchical IMS, network models, and Codd's relational model that led to Oracle.

Key Takeaways

  • Databases predate computers, but computers enabled their automation and scale.
  • The shift from hierarchical and network models to the relational model made data management more flexible and efficient.
  • Codd's relational concept, initially resisted by IBM, became the foundation of commercial database success led by Oracle.

What the video covers

  • Computer databases let people interact with information easily in daily life, from prescriptions to product inventories.
  • Databases existed long before computers; a Sumerian tablet indexing medical prescriptions is an early example.
  • Early computer databases used a flat file model, a simple sequential list that required searching from the start every time.
  • IBM's IMS in the mid-1960s used a hierarchical tree model, where parent nodes point to child nodes, and NASA used it for Lunar Lander drawings.
  • GE's Charles Bachman developed a more flexible network model in which child nodes could have multiple parents.
  • Edgar Codd, an IBM researcher, proposed the relational model, organizing data into tables connected by matching data fields rather than pointers.
  • The relational approach made data easier to access, merge, and change, and was automated by computers.
  • UC Berkeley's Ingres project used Codd's ideas to create a relational database made freely available for a modest fee.
  • IBM took until 1975 to produce its own relational database, while commercial products built on the concept emerged.
  • Larry Ellison, with Bob Miner and Ed Oates, founded a company that shipped Oracle in 1979, the first commercially available relational database compatible with IBM systems.

Answers

Questions about this video

What was the main limitation of the flat file database model?

The flat file model stored records as a simple consecutive list, so every search had to begin at the start and proceed sequentially. This made searching and maintaining large volumes of records slow.

How did the relational model differ from the hierarchical and network models?

The relational model organized data into simple tables connected by matching data fields, rather than using parent-child pointers. This eliminated the need to maintain pointers, making data easier to access, merge, and change.

Why did Oracle become successful in the database market?

Oracle adopted Edgar Codd's relational database concept early and shipped the first commercially available relational database compatible with IBM systems in 1979. IBM had been slower to adopt the concept, which allowed Oracle to sell to IBM's customers and grow into one of the largest software companies.

Full Transcript — Download SRT & Markdown

00:14
Speaker A
our lives are flooded with all kinds of information. Thanks to computer databases, we interact with information easily and seamlessly on a daily basis.
00:41
Speaker A
prescriptions ship manifests card catalogs and product inventories are also databases the computer allowed databases to be automated early computer databases followed a flat file model a simple consecutive list of Records but this format required that computers begin every search at the start of the
01:04
Speaker A
Databases were used to organize information long before the computer age. This Sumerian tablet is an index of medical prescriptions.
01:28
Speaker A
system IMS this tree like structure was parent notes pointing to child nodes was successfully used by NASA to manage drawings for the lunar lander a more flexible network model was developed by g e--'s charles bachmann now child nodes
01:50
Speaker A
Ship manifests, card catalogs, and product inventories are also databases.
02:07
Speaker A
of data into simple tables of related information there were no pointers to maintain because tables are connected only by having matching data fields this made it easier to access merge and change data see JDate an IBM instructor and author was convinced that the
02:26
Speaker A
The computer allowed databases to be automated.
02:54
Speaker A
UC Berkeley used Kahn's idea to create the ingress relational database making it freely available for a modest fee several companies used it as the basis for successful commercial products [Music] it took until 1975 for IBM to produce an
03:16
Speaker A
Early computer databases followed a flat file model, a simple consecutive list of records. But this format required that computers begin every search at the start of the list and search sequentially.
03:43
Speaker A
of his own money to start a software company with Bob miner and add oats their mission developed and sell the first commercially available relational database compatible with IBM system are their flagship product was Oracle which shipped in 1979 the first version ran on
04:04
Speaker A
This was a slow way to search and maintain large volumes of records. New, faster databases were needed that were expandable, reliable, and efficient.
04:26
Speaker A
too late for IBM to dominate the mini computer market and Oracle was already selling to IBM's customers Oracle had adopted Ted Cod's relational database concept early when his own employer IBM wouldn't and Oracle became one of the largest software companies in the world
04:46
Speaker A
In the mid-1960s, IBM used a hierarchical model for their Information Management System, IMS. This tree-like structure, where parent nodes point to child nodes, was successfully used by NASA to manage drawings for the Lunar Lander.
05:12
Speaker A
[Music] you
Topics:database historyrelational databaseflat file modelhierarchical modelIBM IMSOracleEdgar Coddnetwork modelIngresdata management

Get More with the SozAI App

Transcribe recordings, audio files, and YouTube videos — with AI summaries and speaker detection. 30 minutes free.

Or transcribe another YouTube video here →