Hi team: Currently Spark is more and more widely used in hadoop plantform . Should we support `Spark` and `Spark Streaming` job ?