Architectural Anti-patterns for Data Handling

Data: Roulette
Location: C123
Average rating: *....
(1.88, 8 ratings)

This presentation evolved from a serie of observations on real world projects. Initially used as a NoSQL introduction it got to a point where a collection of anti-patterns is shown explaining that is not the case that RDBMS are bad – is the case that sometimes they are badly applied.
Bonus point that nowadays we have a lot of options, from document-based databases to message queues, that enables us to better organize our code.

Map/Reduce, Cache, Indexing and other interesting topics are commented on the way to understanding why there is no silver bullet regarding data handling. Most of the time the fact that systems degrade are not taken into account when deploying your own job scheduler polling a table or when you store logs into a table.

This is not a tentative NoSQL evangelization – because there is no single fit for a project, specially with big data around, but more of an informal comment on my experiences and a catalogue of really interesting non-orthodox ways of perusing a RDBMS .

Photo of Gleicon Moraes

Gleicon Moraes

Gleicon Moraes is director of data engineering at Gleicon loves infrastructure for data, moving large volumes through distributed messaging systems, and databases. He uses Python, Go, and Erlang and focuses on distributed systems, nonrelational databases, and OSS.

Comments on this page are now closed.


Mitsuharu Hadeishi
07/26/2011 8:44am PDT

The technical level of this session was far below what I was expecting from this conference.