Hello, I have a HIVE over HBASE setup using two decent sized machines. I put 1 row of sample data and a count(1) query takes roughly 10 minutes. A select * of that 1 row runs in 0.2 seconds
I am using latest production cloudera dist (CDH4) and followed along https://cwiki.apache.org/confluence/display/Hive/HBaseIntegration http://pastebin.com/DQ279b7b (select count(1) - 10 minutes) http://pastebin.com/vamHTknc (select * , sub-second, fast enough) I understand this isn't meant for interactive query, but 10 minutes is way too slow even if we include all the time to load all the code/jar/anything. Where do I start looking for any configuration issues? Thanks -- Harry
