Another option is Apache Beam. We use it quite extensively. There are a few
options for Clojure wrappers (we use datasplash), and beam has libraries for a
number of popular languages.
Kind Regards,
Dom Parry
On 10 Jul 2020, 08:22 +0200, Alex Ott , wrote:
> From Spark perspective, I would
>From Spark perspective, I would really advise to use Dataframe API as much
as possible, including the Spark Structured Streaming instead of Spark
Streaming - the main reason is more optimized execution of the code because
of all optimizations that Catalyst is able to make. But I really don't see
