I've been reading a lot of documentation trying to find a way to implement a Postgres database on CODA and get it distributed that way. Unfortunately, I found out that CODA apparently doesn't support concurrency over databases, so i came up with an idea of using files, each file containing a single record, so the 2PL locks could be applied to a certain "record-file". This project will use a global directory with the information about the "fragments" and their location. A fragment would consist of a collection of record-files in a coda volume(directory). So my question would be: ¿Is there a way to make a single record dump of a given table to a file?, the fact is that I need an horizontal fragmentation, record by record, so ¿how would you do it? I know this might seem a little nuts, but i would like to get your views on this idea. Please tell me if I'm just wasting my time hehehehehe :) Alex Rivera. PD: please forgive any mispelled words... Alejandro Rivera Polanco Base de Datos 1131300 (ext. 3375) - 1131375 arivera@xxxxxxxxxxxx Grupo Posadas Te invitamos a conocer el nuevo Fiesta Inn Reynosa, donde descansa la gente que trabaja.
