COMMON: (OpenTSDB#1453)

- Use DIFF encoding by default
- Verify DATA_BLOCK_ENCODING variable

Signed-off-by: Chris Larsen <[email protected]>

Author: ZephyrGuo

src/create_table.sh

@@ -19,7 +19,10 @@ BLOOMFILTER=${BLOOMFILTER-'ROW'}
 COMPRESSION=${COMPRESSION-'LZO'}
 # All compression codec names are upper case (NONE, LZO, SNAPPY, etc).
 COMPRESSION=`echo "$COMPRESSION" | tr a-z A-Z`
-DATA_BLOCK_ENCODING=${DATA_BLOCK_ENCODING-'NONE'}
+# DIFF encoding is very useful for OpenTSDB's case that many small KVs and common prefix.
+# This can save a lot of storage space.
+DATA_BLOCK_ENCODING=${DATA_BLOCK_ENCODING-'DIFF'}
+DATA_BLOCK_ENCODING=`echo "$DATA_BLOCK_ENCODING" | tr a-z A-Z`
 TSDB_TTL=${TSDB_TTL-'FOREVER'}
 
 case $COMPRESSION in
@@ -29,6 +32,13 @@ case $COMPRESSION in
     ;;
 esac
 
+case $DATA_BLOCK_ENCODING in
+  (NONE|PREFIX|DIFF|FAST_DIFF|ROW_INDEX_V1)  :;; # Know good
+  (*)
+    echo >&2 "warning: encoding '$DATA_BLOCK_ENCODING' might not be supported."
+    ;;
+esac
+
 # HBase scripts also use a variable named `HBASE_HOME', and having this
 # variable in the environment with a value somewhat different from what
 # they expect can confuse them in some cases.  So rename the variable.