ya, this access pattern doesn't really ring true for cassandra, at least imo
jesse -- jesse mcconnell jesse.mcconn...@gmail.com On Wed, Mar 17, 2010 at 14:14, Ned Wolpert <ned.wolp...@imemories.com> wrote: > I could be wrong, but I would say that even if the thrift syntax gave you a > 'get/set' you want, you would still have the 'inconsistency' risk primarily > because of the Eventually Consistency > (http://www.allthingsdistributed.com/2008/12/eventually_consistent.html) as > implementation in Cassandra. (No concept of a database lock, transaction, > mutex, etc... right?) (Someone please correct me if I'm wrong about this) > > 2010/3/17 Juan Manuel García del Moral <juanman...@taringa.net> >> >> So I would have to retrieve (client.get()) the value, then increment it >> and update it (client.set()) again? with the inconsistency risk this two >> operations imply... >> >> annoying >> >> thanks for your help >> >> >> >> >> 2010/3/17 Jesse McConnell <jesse.mcconn...@gmail.com> >>> >>> afaik, nope >>> >>> -- >>> jesse mcconnell >>> jesse.mcconn...@gmail.com >>> >>> >>> >>> 2010/3/17 Juan Manuel García del Moral <juanman...@taringa.net>: >>> > Hello >>> > >>> > I would like to know if there is any method available to update >>> > (increment/decrement) INTEGER values in Columns/SuperColumns, and how I >>> > can >>> > use this method thru the C++ API >>> > >>> > Thanks >>> > >>> > Juan >>> > >>> > >> > > > > -- > Virtually, Ned Wolpert > > "Settle thy studies, Faustus, and begin..." --Marlowe >
