The PublishKafka processor series is where we really dug into the
details of the Kafka API and build the processors in a manner far more
aligned to the Kafka style. They work well. The PublishKafkaRecord
processor though is *really* nice in terms of performance and it means
it is part of a format and schema aware chain of processing. I'd
start there for maximum performance.
On Thu, Jul 27, 2017 at 4:56 PM, Laurens Vets <[hidden email]> wrote:
> Hi list,
> Simple question really, which Kafka processor should I use to stream items
> into Kafka, more specifically, Kafka as it is used in Apache Metron 0.4.0.
> The sources are very small (1KB) JSON messages.