Re: What data access methods Informix have ?

On Fri, 17 Oct 1997, Jiang Lu wrote: > In Ingres, for each table, it may be exist in four types of data orgnisations, > their heap, hash, isam, btree. In addition, indexes can be created for other > attributes as one of these four types. > > I am reading an Informix book now. it does not address this issue very clearly. > > can any one tell me : > > what type data orgnisations Informix can have? > what indexing methods could be build for each table ? I'm not clear that data organisation is distinct from the indexing methods. On the assumption that it is is, then the data is not specifically organized in any particular layout; new rows are added where there is space in the data structures (where rows have been deleted or at the end if no vacant slots exist, roughly). You can create a clustered index on a table which then organizes the data so that the rows are stored in the same order as the clustered index, but this organization tends to break down over time as rows are deleted and new rows added, and you can only have one clustered index at any given time. In pre-IUS systems, the indexes are B-tree (B+-tree) indexes and that is all -- unless XPS supports bitmap indexes. In IUS, you can write your own indexing code, if you are so inclined. I believe it also supports some extra native index methods. Yours, Jonathan Leffler (johnl@informix.com) #include <witticism.h>