In production at Linkedin, we use a batch size of 200, which has worked
pretty well.

If the producer shuts down, it flushes out the messages in the internal
queue. If it crashes, it does not get a chance to do that and there is no
good alternative to this.

In memory.


NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB