I have a situation where we have constraint on the number of processes we can use and hence, is it possible to load kafka server within another Java process? Looking at the kafka.kafka it seems that I can create a Java object that does most of what Kafka.kafka.main() do. Any thoughts if this will work? What are some of the things I need to watch out? And is there another better way to do this? Thank you for your time.
As Jay says, this is definitely possible. We also do it at Metamarkets.
The one thing that we found odd when doing it though is that you cannot (or couldn't with the version we are using, I'm not sure about 0.8) embed a producer that produces for an embedded broker. You instead have to have your producer talk over the "network" to the broker even though it is running inside the same process. We haven't run into any issues doing this yet, but it's something to be aware of.
On Thu, Feb 14, 2013 at 4:39 PM, Jay Kreps <[EMAIL PROTECTED]> wrote:
Consumer + broker embedding can be done, I believe. It's just producer+broker embedding that they don't like :).
On Thu, Feb 14, 2013 at 4:51 PM, Jamie Wang <[EMAIL PROTECTED]> wrote:
NEW: Monitor These Apps!
Apache Lucene, Apache Solr and all other Apache Software Foundation projects and their respective logos are trademarks of the Apache Software Foundation.
Elasticsearch, Kibana, Logstash, and Beats are trademarks of Elasticsearch BV, registered in the U.S. and in other countries. This site and Sematext Group is in no way affiliated with Elasticsearch BV.
Service operated by Sematext