Data Distribution for Data Warehouse

I'm setting up a somewhat generic data warehouse consisting of up to 100 tables with sizes ranging from 100 MB to 10GB; total diskspace usage may hit 1TB. What are the general rules I should follow to determine chunk size, number of dbspaces, table distribution and fragmentation approach? What (Informix-specific) references should I consult? Using IDS 10.00.FC6 (eventually migrating to 11.7) on RHEL4.

On Jul 20, 5:00 pm, red_valsen <red_val...@yahoo.com> wrote: > I'm setting up a somewhat generic data warehouse consisting of up to > 100 tables with sizes ranging from 100 MB to 10GB; total diskspace > usage may hit 1TB.  What are the general rules I should follow to > determine chunk size, number of dbspaces, table distribution and > fragmentation approach?  What (Informix-specific) references should I > consult? > > Using IDS 10.00.FC6 (eventually migrating to 11.7) on RHEL4. Red_valsen, What do you wanna do ? Is possible to create a dbspace to only one table. Tables with plenties of data, is possible to create fragment. Some tables can be created in external disks. Don't forget to consider backup strategies. With ontape, is possible to restore just one dbspace instead of restoring all of them. Informix publib: http://publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp Administration: http://publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp?topic=/com.ibm.admin.doc/admin23.htm