Timeline

18 moments that made big data what it is. Each one is a publication, a first or a decision that can be looked up.

1. 1890Herman Hollerith’s punched-card machines tabulate the United States census; the previous count had taken most of a decade by hand.
2. 1936A magazine poll of 2.4 million people miscalls the United States presidential election.
3. 1967Gene Amdahl states the limit on speed-up from parallel processing.
4. 1970Edgar Codd publishes the relational model of data. Burton Bloom publishes the Bloom filter.
5. 1985Jeffrey Vitter publishes “Random Sampling with a Reservoir”, the standard account of reservoir sampling.
6. 1986Michael Stonebraker publishes “The Case for Shared Nothing”.
7. 1992Bill Inmon publishes Building the Data Warehouse.
8. 2000Eric Brewer presents the conjecture that becomes the CAP theorem; it is proved two years later.
9. 2001Doug Laney describes volume, velocity and variety.
10. 2003The Google File System paper is published.
11. 2004The MapReduce paper is published.
12. 2006Hadoop begins as an open-source project. The Bigtable paper is published. Differential privacy is defined.
13. 2007The Dynamo paper is published. HyperLogLog is published.
14. 2008Narayanan and Shmatikov publish their demonstration, first circulated in 2006, that a large “anonymous” dataset can be re-identified.
15. 2010Spark is first described. The Dremel paper is published. The term “data lake” appears.
16. 2011Kafka is released as open source.
17. 2014Lazer and colleagues publish their analysis of why a search-based flu estimate failed.
18. 2018The European General Data Protection Regulation begins to apply.