Pig and Hadoop are implemented in Java. From writing a lot of unit
tests, I'd say that almost all of the classes will work correctly on
Windows but if you stack up Hadoop servers on Windows, something is
going to go wrong somewhere. (I've tried it)
A big part of it is that the cost of a Windows license is reasonable
for a developer who has one machine (if you like Windows) but adds up
to a lot of if you have a cluster of 100 machines.
I do my development work on a Windows machine, but I actually run my
development cluster in Ubuntu Linux under VirtualBox. My production
clusters are in Amazon Elastic Map/Reduce at the moment.
You submit Pig and Hadoop jobs with a client-server protocol so it
ought to be straightfoward to build some system in C# that launches a
job, and then you can implement a front end with WPF, Silverlight,
Visual Studio and those sorts of tools.
On Mon, Feb 17, 2014 at 1:20 PM, Ilia Chinkare <[EMAIL PROTECTED]> wrote:
Expert on Freebase, DBpedia, Hadoop and RDF
(607) 539 6254 paul.houle on Skype [EMAIL PROTECTED]