In order to make it easier to import/export data between Vespa and Hadoop/HDFS, should we consider writing a Vespa connector for Apache Sqoop? It could allow an even easier way to move data between the two than using the current feeder for Hadoop even with a nice UI: http://gethue.com/move-data-in-out-your-hadoop-cluster-with-the-sqoop/.
Theoretically, if we have a connector that supports both "from" and "to", this could also be a straight-forward way to move data between Vespa instances?
In order to make it easier to import/export data between Vespa and Hadoop/HDFS, should we consider writing a Vespa connector for Apache Sqoop? It could allow an even easier way to move data between the two than using the current feeder for Hadoop even with a nice UI: http://gethue.com/move-data-in-out-your-hadoop-cluster-with-the-sqoop/.
FYI: @lesters and @bratseth