While I interpret Evan as arguing against using RDB terms like row and column, I would favor keeping those terms. Cassandra's data model is typically *initially* described as a table--without relational aspects!--and then the distinction of its storage strategy (column-oriented, mostly, sort of, with qualifiers) is explained. This has helped me understand the similarities and differences quite well, i.e., a very simplistic view of Cassandra is (RDB without relational, column-oriented-ish, extensible column ids, ...). Understanding the data layout makes it natural to understand the scaling and trade-offs.
As a philosophical aside, the "no-sql" meme emphasizing the exposure of how data is actually stored is a great leap forward. We all need to know all these details and what trade-offs are being made. .. Adam On Tue, Aug 18, 2009 at 10:42 AM, Eric Evans<[email protected]> wrote: > On Mon, 2009-08-17 at 15:32 -0700, Mark McBride wrote: >> My first attempt at a revamped data model wiki page is up here >> >> http://wiki.apache.org/cassandra/DataModel2 > > I think you are on the right track. Very nice. > > -- > Eric Evans > [email protected] > >
