I often get asked why is kdb+ used by banks why not XXXX?
Where XXX is hadoop, mongodb,spark…..
I would turn this around and ask what is your particular goal?
– Take the essential use cases you need
– What criteria does that give us for evaluating a technology stack
– Let’s take into account the skill set of our team
– And the projected Total Cost of Ownership
Then under that criteria let’s see what is the best solution.
We’ve previously posted a comparison of different column oriented databases.
http://www.timestored.com/time-series-data/column-oriented-databases
Nowadays I would expand this further to include data stores available in a particular language, data frames in R, pandas in python…
For kdb+ list it’s strengths:
http://www.timestored.com/kdb-guides/kdb-database-intro
- Kdb+ the Database – Column Oriented DB allowing fast timeseries analysis
- q language – fast, interpreted vector based language
- Combined power of kdb+/q – Change how you think!
Now weight up the pros and cons and decide.