History of Relational Databases

In the older days all the data was stored on magnetic tapes, it was necessary that all the data was to be stored in a sequential fashion, so whenever there was a need to update any data, all the old data had to be read and then merge with the new data, resort it to update to the new data.

Over the years down the line, as storage devices became advanced with the introduction of Hard Disk Drives and Solid State Drives, Random Access became key than how the data was sorted and how efficiently the data could be stored.

In pursuit to make data access more easy and cost effective many companies started to research on making databases to create a system where the related data are allocated in a network fashion and data movement between related information is fast and quick. As of this was in the initial stage, there was a lot of vendor lock in, such as IBM machines could only work with IBM databases and so forth, this wasn’t because they wanted to do it, but because they will still in the research phase and no vendor had any common module to create interoperability.

This mandated a requirement for a common standard, this was done in joint effort with National Institute of Standards and Technology (NIST) and the various vendors, this standard came to be known as Structured Query Language (SQL) sql

SQL is the language we use to talk to databases, some commands are CREATE, DROP, DELETE, UPDATE

Here is some terminology

• Database - contains one or more tables
• Relation - contains tuples or attributes
• Tuple/Row - a set of fields that generally represent an object like a person or a music track
• Attribute/Column - one of many possible elements of data corresponding to the object represented by the row.

Some of the Major Database Systems.

• PostgreSQL - 100% Open Source
• Oracle - Large, Commercial, enterprise-scale, very tweakable
• MySQL - Fast and Scalable, Open Source
• SQLServer - From Microsoft very nice.

Timeline

timeline
    title Evolution of Database Systems
    1960s : Hierarchical & Network Models (IMS, CODASYL)
    1970 : Edgar Codd publishes the Relational Model
    1974 : IBM develops System R and SEQUEL (later SQL)
    1979 : Oracle releases first commercial RDBMS
    1986 : SQL becomes ANSI/ISO standard
    1989 : PostgreSQL project begins at UC Berkeley
    1995 : MySQL released as open source
    2000s : Object-relational features, web-scale databases
    2009 : NoSQL movement gains momentum (MongoDB, Cassandra)
    2010s : NewSQL (CockroachDB, Spanner) combines SQL + distributed scale
    2020s : Cloud-native databases, serverless, multi-model

Key Concepts Invented Along the Way

• Relational Algebra — Codd’s mathematical foundation (1970)
• ACID — Transaction guarantees formalized
• Normalization — Codd’s normal forms for eliminating redundancy
• SQL - Structured Query Language — Standardized query language
• NoSQL Databases — Schema-flexible alternatives for web-scale
• CAP Theorem — Brewer’s theorem on distributed trade-offs (2000)
• Distributed Databases — Sharding, replication, consensus protocols