You could  build a custom secondary index where "column=value" is the key and startKey and endKey are column names. Then call get_count() with a SlicePredicate that specifies the startKey and endKey as the start and finish column names. 


On 5 Apr 2011, at 01:45, Donal Zang wrote:

Can we do count like this?
count <cf>[<startKey>:<endKey>] where <column> = <value>
Donal Zang
Computing Center, IHEP 
19B YuquanLu, Shijingshan District,Beijing, 100049
86 010 8823 6018