Hi,

In TestLoadIncrementalHFiles's runTest method, a hfile is created for each
hfile region (between lines 133-138).
http://svn.apache.org/viewvc/hbase/trunk/hbase-server/src/test/java/org/apache/hadoop/hbase/mapreduce/TestLoadIncrementalHFiles.java?view=markup


I couldn't understand the reason for such an action..

It is obvious that SPLIT_KEYS is used for splitting table into regions,
but what is the use of hfile regions?
Even I wrote all keys into one file, bulk load works...
Does writing into different files according to ranges increases the
performance?
And how should I decide range boundary values?


thanks in advance..

Reply via email to