Apache Spark™ is a fast and general engine for large-scale data processing.

Homepage: http://spark.apache.org/

GitHub mirror: https://github.com/apache/spark

Building instructions: http://spark.apache.org/docs/latest/building-spark.html

Some AURs:
https://aur.archlinux.org/packages/apache-spark/
https://aur.archlinux.org/packages/apache_spark/

I suppose this requires Hadoop, so that (and possible other dependencies) should be added as packages as well.