Hello First in foremost , the kafka scalathon project ( https://github.com/nehanarkhede/kafka-scalathon) has been a great help for me in terms of bootstrapping my learning curve on kafka.
There were 2 issues that I wanted to bring to your attention : 1. The partitioner example with Kafka 0.7 does not seem to be working. All the messages seem to be heading toward only one consumer (although 2 consumers were running) 2. An internal LinkedIn repo is reference in the build ( https://github.com/nehanarkhede/kafka -scalathon/blob/master/project/build/ScalathonProject.scala) - " http://esv4-cm01.corp.linkedin.com:8081/artifactory/ext-libraries"; My workaround to this was to reference my local repository (after a publish-local from kafka 0.7.0 build) Would really appreciate if you could let me know if I am missing anything. On Thu, Feb 23, 2012 at 12:49 PM, Neha Narkhede <neha.narkh...@gmail.com>wrote: > Ahmed, > > You can find relevant design details here - > http://incubator.apache.org/kafka/design.html > > The source code files to look at would be under packages kafka.network > and kafka.log > > HTH, > Neha > > On Thu, Feb 23, 2012 at 12:03 PM, S Ahmed <sahmed1...@gmail.com> wrote: > > Can someone point me to how kafka reads/writes to files in such an > > effecient manner? > > > > I remember watching a video where they described it briefly, it was > > basically taking advantage of java nio but that is all I can recall. > > > > Looking for some good articles that go over what makes it performant etc. > > > > BTW, what files in the source should I be looking at to get a better > idea? > > > > TIA. >
