These are simple workflows where writing a simple script is quicker and easier to maintain than using an ETL system. However, my go-to language Python seems unable to keep up, taking about 3 minutes to parse and insert. An example pattern can be seen at https://gist.github.com/2914601 (I have tried a few tuning patterns, this is the fastest I have come up with). The example uses Redis, as an experiment, but not the only target.

I am interested in looking into languages that will be better performing, but come with some of Python's strengths - concise, easy to write, read and maintain code, and a library of relevant libraries and drivers.

I have been reading a bit about FP languages (Haskell/GHC), because of their performance, but I am not sure if the FP paradigm is appropriate for data integration tasks.

Opinions / Advice ?