I tend to use a real cluster so that I can test at a reasonable fraction of scale. I've seen some instances where code that ran 'okay' in aVM failed to perform adequately at scale.
Sent from a remote device. Please excuse any typos... Mike Segel On Apr 14, 2013, at 2:19 AM, Jens Scheidtmann <[email protected]> wrote: > Dear Vjeran, > > > 2013/4/14 Vjeran Marcinko <[email protected]> >> Hi again, >> >> >> >> You actually touched what I'm trying to do here – setup best Hadooop >> development environment. >> [...] >> >> So are there any more hints for me to setup this environment? >> > > In Eclipse you can use the Maven plugin. See here: > http://www.philippeadjiman.com/blog/2009/12/07/hadoop-tutorial-part-1-setting-up-your-mapreduce-learning-playground/ > Works like a charm for me. > > I also have installed cygwin with Xwindows on my windows computer, so that I > can remotely access the linux box and have my dev environment there. > > In Hadoop's subversion, you can also find an Eclipse plugin, but I didn't try > this yet. http://wiki.apache.org/hadoop/EclipsePlugIn > > Best regards, > > Jens > > > >
