Hello: We would like to avoid the development cycle with compiled languages, such as Java. We also want to only do the map portion, not the reduce. We also want to take advantage of statistical packages within Python.
The question is. What are the best options for doing this? Would we just use python as the streaming application (not sure what support python has for HBase). Would we use something like Groovy that allows to use the Java apis and connect to HBase and then pipe the output into python? Seems there could be many ways to do this. Not sure of what is most commonly done and if there is any documentation related to this. Thanks Brian -- View this message in context: http://apache-hbase.679495.n3.nabble.com/Hadoop-Streaming-and-Hbase-tp4050774.html Sent from the HBase User mailing list archive at Nabble.com.
