database

At the beginning when the relational model was introduced, the main critics was that it will not be possible for a system to optimize the plan of a relational query as a human can do.

• 1970: Edgar F. Codd introduces the notion of relational model with "A relational model of data for large shared data banks"
• 1973: Michael Stonebraker releases Ingres with the query language QUEL, it is the ancestor of Postgres (for Post-Ingres) released in 1996
• 1975: IBM publishes System R, a prototype of RDBMS
• 1979: The first version of Oracle is released
• 1983: IBM releases DB2
• 1996: Postgres is released, first as an object-oriented DB

SQL (SEQUEL) was chosen as a standard instead of QUEL, because Stonebraker refused to talk to the standardization committee. The reason is that he things that standards are bad for innovation.

It started in 80s, the point is to stored the object as they are in object oriented programming languages (quite close to JSON/XML-DB in fact).

In the 2000s

In the 2000s

• Google publishes map reduce ideas
• Yahoo releases Hadoop

The point is to giving up ACID to gain performances

• MongoDB
• Cassandra
• Couchbase
• HBase (Apache)

Every NoSQL systems (except key value store) supports SQL 😁

ACID but faster that SQL with distributed systems. Many of these systems did not survive, the sur

• Google Spanner
• TiDB
• CockroachDB

• Spark
• Amazon Redshift
• Presto
• Apache Drill

• Neo4j
• TerminusDB
• JanusGraph