Because for small columns the Java "container" objects are larger than the actual column data.
On Tue, Sep 13, 2011 at 9:38 AM, Radim Kolar <h...@sendmail.cz> wrote: >> flush of Memtable-idx_graphable@915643571 > (4671498/96780112 serialized/live bytes, 59891 ops) > can somebody explain why difference between live and serialized is so big? > Serialized bytes are actual bytes written to sstable? > > -- Jonathan Ellis Project Chair, Apache Cassandra co-founder of DataStax, the source for professional Cassandra support http://www.datastax.com
